PROCEEDINGS of the LINNEAN 7 SOCIETY of NEW SOUTH WALES VOLUME 132 NATURAL HISTORY IN ALL ITS BRANCHES THE LINNEAN SOCIETY OF NEW SOUTH WALES ISSN 0370-047X Founded 1874 Incorporated 1884 The Society exists to promote the cultivation and study of the science of natural history in all its branches. The Society awards research grants each year in the fields of Life Sciences (the Joyce Vickery fund) and Earth Sciences (the Betty Mayne fund), offers annually a Linnean Macleay Fellowship for research and publishes the Proceedings. It holds field excursion and scientific meetings, including the biennial Sir William Macleay Memorial Lecture delivered by a person eminent in some branch of natural science. Membership enquiries should be addressed in the first instance to the Secretary. Candidates for election to the Society must be recommended by two members. Annual volumes from 133 are available on CD free of charge to any library or institution requesting them. Paper copies are also available; contact the Secretary for costs and details. Most back issues from Volume | can be obtained free of charge on the American website: www.biodiversitylibrary.org/bibliography/6525. You can download individual papers from that wonderful site, which has been composed with the full cooperation of the Linnean Society of NSW. OFFICERS AND COUNCIL 2011/2012 President: D. Keith Vice-presidents: D.R. Murray, M.L. Augee ,1.G. Percival, J.P. Barkas Treasurer: 1.G. Percival Secretary: J-C. Herremans _ Council: M.L. Augee, J.P. Barkas, M. Cotton, E.J. Gorrod, M.R. Gray, J-Cl. Herremans, D. Keith, R.J. King, H.A. Martin, E. May, D.R. Murray, P.J. Myerscough, I.G. Percival, J. Pickett, S. Rose, H.M. Smith, B. Welch and K.L. Wilson Editor: M.L. Augee Assistant Editor: B. Welch Auditors: Phil Williams Carbonara The postal address of the Society is: P.O. Box 82, Kingsford NSW 2032, Australia Telephone: (International) 61 2 9662 6196; (Aust) 02 9662 6196 E-mail: linnsoc@1inet.net.au Home page: www.linneansocietynsw.org.au © Linnean Society of New South Wales Cover motif: Opalised fossils from Lightning Ridge; Figure 2 in the paper by Simone Meakin, page 71 this volume. Photographs by Robert A. Smith courtesy of the Australian Opal Centre. EDITORIAL This is the last printed volume of the Proceedings of the Linnean Society of New South Wales. From the first volume in 1877 to 2011, our journal has adapted to many changes in the technology of printing. In 2011 it is now time to make a much more dramatic change — away from print and paper into the electronic age. From volume 133 the Proceedings will be published electronically on the world wide web through e-Scholarship (a division of the University of Sydney Library). Authors in 2011 require rapid publication. Authors of scientific papers require credible publication, which the Linnean Society of NSW will provide by continuing to subjecting all manuscripts to rigorous peer review. The mode of publication, through a secure site that only hosts material from scientific and academic organisations, further ensures credibility. Papers will be put on the Net as soon as they have been reviewed, revised as necessary and accepted by the Society’s publications committee. At the end of each year all papers put on the Net during that year will be “bound” into a numbered volume for permanent archiving by e-Scholarship. That volume will also be available on CD free-of-charge for all our current subscribers around the world and to members of the Linnean Society of NSW. ; One huge advantage of this new mode of publication is that the papers, and hence the information provided by the scientific research therein, will be available world wide free-of-charge. It is the current policy of the Society that access to papers as they are published and archived material at e-Scholarship will remain free. At present and into the foreseeable future all of the Society’s journals from Volume | are available free- of-charge through the American information site biodiversityheritagelibrary.org. This change is not 100% free of paper and print, as existing requirements of the Zoological and Botanic - Nomenclature Commissions force us to distribute a small number of printed copies to selected institutions. This requirement is bound to change in the future. Also, e-Scholarship has the facility to provide (with a charge) printed and bound individual copies of the electronic volumes upon request. This volume contains papers arising from symposium on Geodiversity, Geological Heritage and Geotourism held by the Linnean Society of New South Wales in September 2010. They comprise the first section. The second section contains general papers. WHE LANNEAN SOCH iW SOUTH WALA ISSN 0370-647%X LALSOYYICIS > MW acd a ilies, cones as A onl eteay ys Gil) Ver 2ettil) i? bali seat cele ab coll : j rod wae vitae ey on ese v rat (voiieden all beero) toe 1/05 on ST RE go eae wll E yy one <5 ny “f puns a] veh ges stent ate How ibe silts of ont congas ey ; Ls > ou wi on oleae dan iin ind bivakoeruenl ort: CF! a ‘ ' s inet wii) he L Ve Opi rser i. View Wat moh MU ed si ererngag'J oh to a eoihey Wid wed 1 rif = es oie e" ahs We ore & ‘ wii. Laie ulecutn a 9 aelibers TPO? vecn bsg ad) otis Wad st) moist aie bo oe POSS VRGRY sah 0 m aqronunein | - Re iy an Hilditesrs to's set} Wiareeyigut a Bie Minos MM nc Se aay ‘. LY ee | 2p 4 Oa ey Sa eared ; > dat tle tose desi en tiple MP's et AES «era. dt od | atin Liew yl) eC Yd ali meee se herein "braid ont te t shir tos Maat tin ibe ve acti bs cree ais toraee Vw atk Te Me a < OgA: | Sena ; eaups coq aeimdloy wore sit. 2 “eo. ) 2 * % con'h ‘ae @ « -_. oo © #8 bags on Se a anor a ; Cas Mah ViCTOLAN Seen: i * heamnumuyitbeald Laurie): eeeeenate an) yp et een ARE rt wecatnen saalov MeT ¥ 4 wn sty seni vad O10 wertectied of elt diuod’ wed Yo vioived’ nasanit seb wl bled reroute 4 eT MILA) LTO xettnareD ) covase banaue of F ai: | - ; r 7 ! | ' Rd ft sets : a. BF. Py. Mycesiaeny y _ : 7 ps prada adler rf dhe Sholay ta ARNG 2 Khe Pcheyta de — ane dull } 63 , a Ce eat aS Saar 7 b, areadl” april ate niet i c” ) ees eu > Howie Dese Gene Iniunciniciehyne yen BG,” i ; ani ‘i. 4 Sa a + ' ine ® dammenwe Ratety of New South Walks, eae i, i Co v e PROCEEDINGS of the LINNEAN SOCIETY of NEW SOUTH WALES For information about the Linnean Society of New South Wales, its publications and activities, see the Society’s homepage www.linneansocietynsw.org.au VOLUME 132 May 2011 } | | » a Law HIOOS W3A, id % i v 7 - i rs i ea 4 a = =a i i. ; s% ) | Le NS of ling erie vii alk dine? wal Ww visited me arint A vill ‘uci coisa 157 1 r 1 een / : 7 pat oe Seemed KE yiarod wll O92 , eallivilos Ti \ eI ; , t ; os * 7 ag: Se IO Men (vata na uw Neat oa on! FOREWORD TO PAPERS FROM A SYMPOSIUM ON GEODIVERSITY, GEOLOGICAL HERITAGE AND GEOTOURISM Geotourism, Geodiversity and Geoheritage in Australia — Current Challenges and Future Opportunities ANGUS M. Rosinson! AND IAN G. PERCIVAL2” ‘Leisure Solutions® P O Box 638, Strawberry Hills, NSW 2012, Australia (angus@leisuresolutions.com.au). *Geological Survey of New South Wales, WB Clarke Geoscience Centre, 947-953 Londonderry Road, Londonderry, NSW 2753, Australia (ian.percival@industry.nsw.gov.au). Robinson A.M and Percival I.G. 2011. Geotourism, Geodiversity and Geoheritage in Australia — Current Challenges and Future Opportunities. Proceedings of the Linnean Society of New South Wales 132, 1-4. Geotourism, in addition to its primary role in promoting tourism to geosites, raises public awareness and appreciation of geodiversity. It fosters geoheritage conservation through appropriate sustainability measures and advances sound geological understanding through interpretation. Currently in Australia, geotourism is in its infancy and faces a range of challenges, including lack of awareness and support within the geological professions and varying degrees of acceptance by natural resource managers. Geodiversity on the other hand is now widely appreciated as part of the natural heritage, and is being integrated into government policy concerning the management of national parks and public lands to a degree approaching the stewardship of the native flora and fauna, as greater emphasis is placed on the underlying control of distribution of the living environment by geology and landscape. Conservation of geodiversity and geoheritage is thereby progressing rapidly in some areas, though in others such as the development of geoparks in the Australian context, significant barriers have yet to be surmounted. The recent Symposium on Geodiversity, Geological Heritage and Geotourism, organised by the Linnean Society of New South Wales at Port Macquarie in September 2010, provided an opportunity to discuss these matters from a number of viewpoints, including government, academic and the private sector. Manuscript received 10 November 2010, accepted for publication 20 April 2011. KEYWORDS: experiential tourism, geoheritage, geodiversity, geoparks, geotourism, national landscapes. Given Australia’s heavy reliance on the expertise of geologists and the exploitation of natural resources for wealth creation, it would be logical to assume that the interpretation of geology and landscape feature extensively in the character of Australia’s ‘nature- based’ tourism industry. However, geotourism is at a very early stage of development in Australia, and faces many challenges, ranging from achieving agreement on what the term actually means, to building a support and advocacy base and further to raising awareness amongst Australian domestic travellers. In comparison, appreciation for geodiversity as an essential part of the natural environment is well advanced, and — thanks to Australia’s diverse underlying geology and associated scenic landscapes —many national parks and other public lands protect a broad spectrum of geological heritage sites that are either current or potential foci of geotourism. Natural Heritage, Geoheritage, Ecotourism, and Geotourism Natural heritage is the legacy of natural objects and intangible attributes encompassing the countryside and natural environment, including biodiversity (the variety and distribution of flora and fauna), as well as geodiversity (involving landforms and geology). Geoheritage is exemplified by geological sites of outstanding and sometimes unique scientific and scenic value which enable us to understand the composition of the earth, the internal and external FOREWORD processes that have shaped it, and the evolving flora and fauna that occupied it. Geodiversity and geological heritage are best experienced by visiting natural places, thereby providing the rationale for geotourism, now increasingly considered a_ key driver of ‘experiential’ tourism. Like ecotourism, geotourism is ecologically sustainable tourism with a primary focus on experiencing natural areas that fosters environmental and cultural understanding, appreciation and conservation. Geotourism enables the public to explore the full range of geodiversity, and can be undertaken in a range of places that include geosites, geo-trails, landforms, karst areas and caves, and mine sites. In addition, geotourism can embrace a range of designated areas which include national parks/reserves/urban parks, world heritage areas, ‘national landscape’ areas, and geoparks and paleoparks. The downside of the popularisation of ecotourism in recent years is that the activity itself may progressively destroy the very values that appeal to the ecotourist. This is a continuing problem, particularly now as the greatest impact of mass ecotourism is falling on the most fragile of environments. Thus geotourism must seek to understand its impact on geodiversity and strive to protect and expand geological heritage sites that form in many instances the basis for its existence. GEOPARKS — THE AUSTRALIAN EXPERIENCE A geopark is defined by the United Nations Educational, Scientific and Cultural Organization (UNESCO) as a territory encompassing one or more sites of scientific importance, not only for geological reasons but also by virtue of its archaeological, ecological or cultural value. There are currently 77 global geoparks operating in 24 countries around the world as part of the UNESCO Geoparks International Network. One of the most important aspects of a geopark is the link between the geology and the people, their stories, culture and history that builds a sustainable source of geotourism, brings jobs to rural and indigenous people and in turn, helps protect sites of importance, and promotes geoheritage. Although a geopark has no formal protected lands status (unlike a nature park managed by a government agency), an existing national park or any other designated area may qualify as a geopark, if it has a management plan designed to foster socio-economic development that is sustainable (most likely to be based on geotourism). In addition, the proponents of the geopark must (1) demonstrate methods for conserving and enhancing geological heritage and provide means for teaching geoscientific disciplines and broader environmental issues, and (2) have prepared joint proposals submitted by public authorities, local communities and private interests acting together, which demonstrate the best practices with respect to geoheritage conservation and its integration into sustainable development strategies. Kanawinka Geopark, the first (and currently, only) one in Australia, was declared in 2008. It occupies an area of 26,910 square kilometres spanning nine Shire Council areas in southwestern Victoria and eastern South Australia. The geopark represents the sixth largest volcanic plain in the world with 374 eruption points, with Recent volcanism extending from the Mount Gambier area in South Australia into the Portland (Victoria) shoreline and north as far as Penola and Mount Hamilton. However, Kanawinka Geopark has so far been unable to gain Australian Government approval which would enable UNESCO to assign ‘global geopark’ status. Australian Government Ministers for the Environment and Heritage (EPHC) met in November 2009 and decided that whilst Australian governments support geological heritage, they have significant concerns with the application of the UNESCO Geoparks concept in Australia, especially without government endorsement. Furthermore they determined that existing mechanisms are considered sufficient to protect geoheritage in Australia. In its formal communiqué, the Ministerial Council also requested the Australian Government ask UNESCO to take no further action to recognise any future proposals for Australian members of the Global Geoparks Network, or to further progress Geoparks initiatives within Australia, including that for the Kanawinka Geopark, unless the formal agreement of the Australian Government has first been provided. As a response to the EPHC decision, the author of one of the presentations at a subsequent Linnean Society of NSW Symposium suggested that several other issues need to be addressed before geopark development can proceed any further in Australia. These issues include the following. 1. There are other competing ‘land designation’ systems underpinned by environmental, heritage and tourism values e.g. national parks, world heritage areas, including ‘national landscapes’. 2. The nature of Australia’s political system means that any geopark proposal needs to be accommodated and supported by three levels of government. Proc: Emn. Soc. N_S-W2 13252010 A.M. ROBINSON AND L.G. PERCIVAL 3. Thereisarelatively low profile of geoscience in the Australian community — overshadowed by the strong influence of the Australian mining industry lobby. 4. Apathy amongst the Australian geological community is not helped by the decline in geoscience education and university geology schools in recent years. 5. The geopark concept is not yet embraced or understood by the geological profession. 6. The agricultural/mining industries (which have competing land requirements) are yet to be engaged. 7. The state/territory Geological Surveys and Geoscience Australia are not yet engaging to any significant extent in geopark development and geotourism generally. 8. No government funding programs are available for geopark development. AUSTRALIAN GEOTOURISM — FUTURE OPPORTUNITIES Australia’s National Landscape Program ‘Experiential tourism’ has been captured in the Australia's National Landscapes program (a partnership of Tourism Australia and Parks Australia), where visitors can experience the best of Australia’s natural, cultural and spiritual wonders — to be known as “Experiencescapes.’ These are world-class landscapes distinctive to Australia, and include many geoheritage sites. The National Landscapes program currently includes the following 10 regions: Australian Alps (New South Wales/Victoria), Australia’s Green Cauldron (New South Wales/SE Queensland border region), Australia’s Red Centre (Northern Territory), Australia’s Coastal Wilderness (New South Wales/ Victoria), the Flinders Ranges (South Australia), Kangaroo Island (South Australia), the Great Ocean Road (Victoria), the Greater Blue Mountains (New South Wales), the Kimberley (Western Australia), and West Armhem/Kakadu/Nitmiluk (Northern Territory). Four other regions are also under active consideration viz. Ningaloo-Shark Bay (Western Australia), South Coast (Western Australia), the island of Tasmania, and the Great Barrier Reef (Queensland). Two other areas (i.e. Sydney Harbour and the Wet Tropics area of North Queensland) have been nominated for discussion. Geotourism and Mining Sites Geoheritage A significant feature of geotourism is that it does not require untouched landscapes as its playground. Proce, tinn. Soc. N.Saw.als2, 2011 A great tour can equally be delivered overlooking a man-made excavation, or in a historic mining area e.g. Broken Hill in New South Wales, Chillagoe in North Queensland and the West Coast of Tasmania. Nor does geoheritage potential need to be restricted just to geological features. For example, the Australian Government Department of Environment, Water, Heritage and the Arts has been assessing both the mining and minerals (i.e. economic geology) heritage of Broken Hill from the perspective of the following attributes: (1) Broken Hill’s prominent role in Australia’s mining history; (2) its role in the development of innovative mining and metallurgical practices; (3) as the place where safe working practices and workers’ legislation was first developed for miners; (4) for its well-known mineralogical diversity; and (5) for its importance for the associations with many individuals who have played a prominent role in the Australian mining industry. Geoheritage, Geotourism and the Geological Profession In July 2010, at the Australian Earth Sciences Convention (AESC 2010), a workshop was organised in collaboration with the Geological Heritage Standing Committee of the Geological Society of Australia, The AusIMM, and the Australian Department of the Environment, Water, Heritage and the Arts, to explore the interface between the issues relating to geoheritage and the emerging area of geotourism. Of nine formulated workshop conclusions, the following points are considered particularly relevant in the context of understanding opportunities for geotourism development in Australia. 1. Given the broad range of concepts encompassed by and related to geoheritage, there is a need for the geological profession (or more generally the geoscience professions) to engage further with relevant government agencies to improve mutual awareness and understanding, as well as to better coordinate interaction with relevant government agencies. 2. There is a need to make better known to established and prospective geotourism operators and others the availability of various state/territory resources which identify and promote geoheritage sites. This should include information on site suitability for geotourism. 3. There is continuing concern about the lack of understanding both within the geoscience professions and the general community of the differences between the concepts of geoheritage and geotourism. FOREWORD 4. Interest in mining heritage can be expanded to embrace areas of geoheritage pertaining to economic geology (i.e. relating to the minerals industry). Moreover, there is an opportunity to encourage individual mining companies and industry associations to assist with funding aimed at helping in the conservation of geoheritage and to foster higher levels of community awareness through the support of geotourism activities, where practicable. 5. There is an opportunity to foster and promote geotourism initiatives within Australia’s National Landscapes with geological and geomorphological significance, as a model to advancing geotourism and geoheritage considerations in other regions, having particular regard to the recently stated views of the EPHC relating to the advancement of geopark proposals in Australia. LINNEAN SOCIETY OF NSW SYMPOSIUM ON GEODIVERSITY, GEOLOGICAL HERITAGE AND GEOTOURISM — AN OVERVIEW The recent Linnean Society Symposium, held at Sea Acres National Park in Port Macquarie, NSW from 6-10" September, 2010, addressed many of the issues, challenges and opportunities discussed above. The Symposium was co-sponsored by the Geological Survey of NSW (GSNSW), part of I&I NSW, and the Department of Environment, Climate Change and Water NSW (DECCW) through its Karst and Geodiversity Unit. Both organizations provided a number of speakers. Others amongst the 55 registrants came from the Commonwealth Department of Environment, Heritage and the Arts, and from the Tasmanian Department of Primary Industries, as well as researchers from several universities and museums, teachers, private sector tourism operators and ecological consultants, and retired persons active in local geotourism ventures. Several days of talks were interspersed with two day- long field trips to investigate various aspects of the regional geodiversity of the mid North Coast. Abstracts from the papers presented at the Symposium, together with the guide to the field trip localities, were compiled into a book produced for attendees. All abstracts and most presentations given at the Symposium are available for download at —_http://www.dpi.nsw.gov.au/minerals/geological/ info/geodiversity-symposium. A link to this site is available on the Linnean Society’s website www. linneansocietynsw.org.au. Nine of the papers from the Symposium have been submitted for publication in this volume of the Proceedings of the Linnean Society of NSW. Others, which were more concerned with government policy issues, legislation pertaining to geodiversity and geoheritage conservation, or geotourism, are being prepared for publication elsewhere. *(Ian Percival publishes with permission of the Director, Geological Survey of New South Wales]. Proc. Linn. Soc. N.S.W., 132, 2011 SUT Ty Between Geodiversity and Vegetation in South- eastern Australia Davip A. KEITH NSW Department of Environment, Climate Change & Water, PO Box 1967, Hurstville 2220, NSW [david. keith@environment.nsw.gov.au] and Australian Rivers and Wetlands Centre, University of New South Wales. Keith, D.A. (2011). Relationships between geodiversity and vegetation in south-eastern Australia. Proceedings of the Linnean Society of New South Wales 132, 5-26. Geodiversity, the natural range of geological, geomorphological and soil features, is thought to play a key role in the development of Australian ecosystems and evolution of their biota, due to the widespread occurrence of old soils with impoverished nutritional status and the comparatively restricted occurrence of fertile soils. While associations between soils and vegetation characteristics such as scleromorphy were first noted a century ago, modern theories propose evolutionary and ecological processes that shaped Australian flora and vegetation through interactions between soil nutrition, plant functional traits and flammability. Evidence in support of these generalisations comes mainly from site-specific empirical studies, surveys of plant traits and their associations, classification and mapping of land systems, analyses of regional environmental gradients and phylogenetic studies. The extent to which soils and the substrates from which they are derived place constraints on the climatic response of biota has important implications for understanding future responses of the biota to anthropogenic climate change. Yet, worldwide, there are few studies that examine the relative contributions of geological substrates and climate to variation in vegetation properties over extensive sub-continental regions. This study used a spatially explicit approach to examine the relationship between vegetation and geological substrates over New South Wales, a region of 80 million hectares in south-eastern Australia spanning a diverse range of geology, vegetation and climate. It aimed to assess the fidelity of major vegetation types to geological substrates and estimate the overall influence of geodiversity on vegetation composition relative to that attributable to climate. The spatial data were drawn from maps produced by geological and vegetation surveys. Geological maps were re-classified into 16 broad units reflecting textural and mineral characteristics considered likely to be influential on plant growth. They represent a component of total geodiversity related to broad landscape-scale patterns in major bedrock and regolith types. Vegetation maps were re-classified in 16 broad formations reflecting structural, physiognomic and functional characteristics of vegetation, and into a larger number of 99 classes reflecting species composition. The two spatial data sets were analysed to determine the diversity of vegetation types within geological units and fidelity of each vegetation type to each geological unit. The relative influence of climatic co-variables was also examined using spatial surfaces spline-fitted to 30 years of weather station data. The results indicate a strong non-random relationship between vegetation and geodiversity, with most vegetation types restricted to a narrow range of geological substrates. Partial variance analyses indicated that the influence of geodiversity on vegetation composition was stronger than, and largely independent of the influence of climate. Consistent with current theories, sclerophyllous vegetation formations and classes showed a strong association with geological units characterised by low levels of mineral nutrients. It was concluded that landscape patterns of geodiversity are likely to place significant constraints on the response of native vegetation to future climate change. Manuscript received 7 November 2010, accepted for publication 16 March 2011. KEYWORDS: climate change, environmental gradient, floristic composition, geological map, gradient analysis, landscape biogeography, Old Climatically Buffered Infertile Landscapes - OCBIL, sclerophyll, soil fertility, vegetation-soil relationships, vegetation classification, vegetation map. RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION INTRODUCTION Geodiversity is the natural range (diversity) of geological(rocks, minerals, fossils), geomorphological (land form, processes) and soil features. It includes their assemblages, relationships, properties, inter- pretations and systems (Gray 2004). Many of the earliest phytosociological studies in south-eastern Australia noted the association between plants and the geological substrates and land forms on which they grow (McLuckie & Petrie 1927, Pidgeon 1937, Fraser & Vickery 1939, Crocker 1944, Beadle 1948). The soils produced by weathering of these substrates in particular geomorphic settings vary greatly in levels and proportions of mineral nutrients, and in their textural and structural characteristics that govern their capacity to retain moisture and conduct subterranean oxygen. Different plant species vary in their ability to extract these three essential resources and to tolerate extreme levels of supply. Species that share similar ranges of tolerance to soil-related resources may therefore be expected to co-occur within communities on particular groups of geological substrates, at least in cases where there is vertical concurrence between soils and the underlying substrate. At biogeographic scales, geological substrates are thought to play a key role in the development of Australian ecosystems and evolution of their biota, due to the widespread occurrence of old soils with impoverished levels of nutrients and the comparatively restricted occurrence of fertile soils. Diels (1906) first noted a distinction between sclerophyll vegetation associated with sandy soils and ‘savanna’ vegetation associated with ‘favourable’ soil conditions. Andrews (1916) later proposed a connection between sclerophylly (a syndrome typified by small, thick leaves with thick cuticles and abundant sclerotic tissue) and soil nutrition, particularly nitrogen and calcium. By mid-century, empirical evidence was emerging that physiological and morphological traits related to uptake of phosphorus (Beadle 1953), acquisition of nitrogen (Hannon 1956) and accumulation of aluminium (Webb 1954) were closely associated with acidic, nutrient-deficient soils that occur in certain parts of the Australian continent. Modern theories propose evolutionary and ecological processes that shaped Australian flora and vegetation through interactions between soil nutrition, plant functional traits and flammability (Beadle 1966, Orions & Milewski 2007, Hopper 2009, Lambers et al. 2010). Evidence in support of these generalisations comes mainly from empirical site-specific or autecological studies (Beadle 1954, Webb 1954, Lamont 1982, Shane & Lambers 2005), surveys of plant traits and their associations (Gill 1975, Lambers et al. 2010), the classification and mapping of land systems — areas of recurring patterns in topography, soils and vegetation (Christian 1952), analyses of local and regional environmental gradients (Myerscough & Carolin 1986, Keith & Sanders 1990) and phylogenetic studies (Johnson & Briggs 1975, Crisp et al. 2004). Sclerophyll elements of Australian vegetation appear to have originated on pockets of oligotrophic soils in the Cretaceous (Hill et al. 1999, Crisp et al. 2004), yet they did not diversify and rise to dominance until 25 — 10 million years ago. Their expansion and diversification was at the expense of mesic forest vegetation, and coincided with climatic cooling, drying and increased seasonality as separation of Australia and South America from Antarctica initiated circum-polar oceanic currents (Crisp et al. 2004). Expansion and diversification of the Australian arid flora occurred more recently, 5 — 2 million years ago, as the continent moved still further north and experienced extreme wet-dry glacial cycles (Crisp et al. 2004). Fires also became prominent periodically and influential on vegetation during this period (Keeley & Rundel 2005). Yet strong edaphic patterns apparently persisted through this history and remain in the contemporary vegetation (Hopper 2009) and, despite major extinctions and radiations associated with climatic upheavals, there is evidence of biome conservatism in a large majority of lineages (Crisp et al. 2009). While both soils and climate play prominent roles in theories of the evolutionary history of species within Australian vegetation (Crisp et al. 2004, Orians & Milewski 2007, Hopper 2009), their historical inter-relationships are poorly understood. How much did soils constrain the historical responses of vegetation to climate change? This question seems crucial to understanding the future response of biota as anthropogenic climate change unfolds, yet the spatial dimensions of historical biomes are difficult to quantify and hitherto remain largely unexplored. Insights can possibly be gained by using spatially explicit data to study contemporary relationships between vegetation, soils and climate, yet this has not been examined at the level of assemblages over extensive bioregional scales. This study investigated the relative influence of geological substrates and climate on the biogeography of vegetation at a sub-continental spatial scale. Its aims were: 1) to determine the fidelity between major vegetation types and geological substrates; and ii) to estimate the relative contribution of substrates and climatic variables to variation in species composition of the vegetation. The study was carried out across Proc. Linn. Soc. N.S: W., 132,201 D.A. KEITH New South Wales, a region of 80 million hectares in south-eastern Australia that currently encompasses a diverse range of vegetation, geology and climate. Historically, the region underwent profound climatic upheaval since the appearance of flowering plants in the Cretaceous. A spatially explicit approach was applied using map data synthesised from extensive vegetation and geological surveys carried out within the region (Keith 2004, Stewart et al. 2006). The two spatial data sets were first analysed to assess the fidelity between vegetation types and geological units. A second analysis was carried out to assess the relative influences of geological substrate and climatic factors on the species composition of vegetation. Both analyses were carried out on broad scale classifications reflecting on respective maps the broad structural, physiognomic and functional characteristics of vegetation and the broad textural and mineral characteristics of geological substrates. METHODS Vegetation map The vegetation map was assembled from 105 source maps covering various subregions within New South Wales. These source maps employed a range of different vegetation classifications, varying spatial scales and overlapped with one another spatially to varying degrees. To simplify and standardise the source maps to a common format, the legend of each was re-classified into the vegetation formations and classes described by Keith (2004). In this classification 16 broad formations and sub-formations represent variation in structural, physiognomic and functional characteristics of vegetation (Table 1), and 99 vegetation classes nested within them represent variation in vascular plant species composition (Keith 2004). Classes are not strictly nested within formations, as structure may vary considerably within compositionally defined units, however, classes were assigned to the vegetation formation representing the most commonly expressed structural, physiognomic and functional features of a mature stand (Keith 2004). The 105 source vegetation maps were ranked according to their relative reliability based on assessments of classification skill, thematic and spatial resolution and currency using a protocol adapted from the one described by Keith & Simpson (2006). After standardising their spatial projections to the Australian Geodetic Datum 66 and Lamberts Conformal projection in a geographic information system, the source maps were merged sequentially Proc. Linn. Soc. N.S.W., 132, 2011 so that features of the most reliable source map were displayed in preference to those of other maps wherever there were spatial overlaps (Keith 2004). A similar procedure was used to prepare a mask of extant native vegetation by reclassifying each legend category of each source map as native or non-native and then merging according to the currency of remote imagery from which each source map was derived. The mask was then applied to the composite map to derive an updated version (v3.0) of the vegetation map of NSW and the ACT prepared earlier by Keith (2004) (Fig. 1). Geological substrate map A map of geological substrates was derived from spatial data on the surface geology of Australia (Stewart et al. 2006) which, for New South Wales, was based primarily on mapping prepared by the Geological Survey of New South Wales (GSNSW, http://www.dpi.nsw.gov.au/minerals/geological/ geological-maps). This dataset shows the distribution of 785 geological units (mainly at Geological Formation level) and was generalised largely from the state digital geology dataset as at 2003, comprising the GSNSW 1:100 000 and 1:250 000 geological map series. Some areas in the north-east of the state (eg: Moree, Inverell, Tamworth, Manilla) and central-west (eg: Goulburn, Lake Cargellico) were re-compiled from more recent data sourced from GSNSW in 2004-5 (Stewart et al. 2006). The basement geology of the Broken Hill region was compiled from 1996 1:500 000 scale data from the national, NSW and South Australian geological surveys. Mapping for the Murray Basin region was compiled from 1991 1:1 000 000 scale data from the Australian Geological Survey Organisation (AGSO). To compile a seamless state dataset, the original map data were edited along the edges of source datasets (edge-matching), which varied in age and spatial scale of compilation. Adjustments to some older geological data were made using geophysical data interpretation where particularly poor edge-matching or spatial accuracy (+ 1 km) was identified in the source data (Stewart et al. 2006). The spatial data prepared by Stewart et al. (2006) generally did not distinguish contrasting depositional environments of unconsolidated Quarternary sediments along the coast. To resolve this deficiency, the coastal portion of Stewart’s map was overlain by more recent 1:25 000 scale mapping of coastal Quarternary geology, which discriminated a further 48 units of sediments with alluvial, estuarine and coastal barrier depositional systems (Troedson & Hashimoto RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION Formation ainforests Wet sclerophyll forests (shrubby subformation) Wet sclerophyll forests (grassy subformation) Grassy woodlands Grasslands Dry sclerophyll forests (shrub/grass subformation) Dry sclerophyll forests (shrubby subformation) Heathlands Alpine complex Freshwater wetlands Forested wetlands Saline wetlands Semi-arid woodlands (grassy subformation) Semi-arid woodlands (shrubby subformation) Arid shrublands (chenopod subformation) Arid shrublands (acacia subformation) Table 1. Vegetation formations of New South Wales. Description orests of broad-leaved mesomorphic trees, with vines, ferns and palms. Includes Cunoniaceae, Sapindaceae, Monimiaceae, Lauraceae, Meliaceae, Myrtaceae, Apocynaceae, Rubiaceae, Aspleniaceae, Dryopteridaceae. Coast and tablelands in mesic sites on fertile soils. Tall forests of scleromorphic trees (typically eucalypts) with dense understories of mesomorphic shrubs, ferns and forbs. Includes Myrtaceae, Rubiaceae, Cunoniaceae, Dryopteridaceae, Blechnaceae, Asteraceae. Relatively fertile soils in high rainfall parts of coast and tablelands. Tall forests of scleromorphic trees (typically eucalypts), with grassy understories and sparse strata of mesomorphic shrubs. Includes Myrtaceae, Poaceae, Euphorbiaceae, Fabaceae, Casuarinaceae and Asteraceae. Coast and tablelands in high rainfall regions on relatively fertile soils. Woodlands of scleromorphic trees (typically eucalypts), with understories of grasses and forbs and sparse shrubs. Includes Myrtaceae, Poaceae, Asteraceae, Epacridaceae and Pittosporaceae. Rolling terrain with fertile soils and moderate rainfall on the coast, tablelands and western slopes. Closed tussock grasslands with a variable compliment of forbs. Includes Poaceae, Asteraceae, Fabaceae, Geraniaceae and Chenopodiaceae. Fertile soils of the maritime zone, tablelands and western floodplains. Forests of scleromorphic trees (typically eucalypts), with mixed semi-scleromorphic shrub and grass understories. Includes Myrtaceae, Poaceae, Asteraceae, Ericaceae, Dilleniaceae and Fabaceae. Moderately fertile soils in moderate rainfall areas of the coast, tablelands and western slopes. Low forests and woodlands of scleromorphic trees (typically eucalypts), with understories of scleromorphic shrubs and sparse groundcover. Includes Myrtaceae, Proteaceae, Ericaceae, Fabaceae and Cyperaceae. Regions receiving high to moderate rainfall on the coast, tablelands and western slopes. Dense to open shrublands of small-leaved scleromorphic shrubs and sedges. Includes Proteaceae, Fabaceae, Myrtaceae, Casuarinaceae and Cyperaceae. High rainfall regions of the coast and tablelands on infertile soils, often in exposed topographic positions. Mosaics of herbfields, grasslands and shrublands. Includes Ericaceae, Asteraceae, Gentianaceae, Ranunculaceae, Poaceae and Cyperaceae. High, snow-prone parts of the southern ranges. Wet shrublands or sedgelands, usually with a dense groundcover of graminoids. Includes Cyperaceae, Restionaceae, Juncaceae, Haloragaceae, Polygonaceae, Ranunculaceae and Myrtaceae. Throughout NSW on peaty, gleyed or periodically inundated soils with impeded drainage. Forests of scleromorphic trees (eucalypts, paperbarks, casuarinas) with sparse shrub strata and continuous groundcover of hydrophilous graminoids and forbs. Includes Myrtaceae, Cyperaceae, Ranunculaceae, Blechnaceae, Poaceae. Floodprone plains and riparian zones principally along the coast and inand rivers. Low forests, shrublands and herbfields of mangroves, succulent shrubs or marine herbs. Includes Verbenaceae, Chenopodiaceae, Juncaceae and Poaceae. Coastal estuaries and saline sites of the western plains. Open woodlands of scleromorphic trees (eucalypts, acacias), with open understories mostly of chenopod shrubs, usually with strong representation of perennial and ephemeral grasses and forbs, including many ephemeral species. Includes Myrtaceae, Fabaceae, Chenopodiaceae, Asteraceae, Poaceae and Polygonaceae. Low-moderate rainfall regions of the near western plains on clay soils, including infrequently flood-prone sites. Open woodlands of scleromorphic trees (eucalypts, acacias, casuarinas), with open understories of xeromorphic shrubs, grasses and forbs, including many ephemeral species. Includes Myrtaceae, Cupressaceae, Fabaceae, Myoporaceae, Sapindaceae, Asteraceae, Poaceae and Acanthaceae. Low-moderate rainfall regions of the near western plains, including infrequently flood-prone sites. Open shrublands of chenopod shrubs, with perennial tussock grasses and ephemeral herbs and grasses. Includes Chenopodiaceae, Asteraceae, Aizoaceae, Fabaceae and Poaceae. Low rainfall regions of the far western plains. Open shrublands of xeromorphic shrubs, hummock or tussock grasses and ephemeral herbs and grasses. Includes Fabaceae, Proteaeceae, Myoporaceae, Asteraceae, Casuarinaceae and Poaceae. Sandy or rocky landscapes in low rainfall regions of the far north-western plains. Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH 0 50100 es Kilometres 200 300 400 Vegetation Formations ee Rainforests za] Wet sclerophyll forests (Grassy subformation) FEE) Wet sclerophyll forests (Shrubby subformation) PASS Grassy woodlands S| Grasslands | | Alpine complex BE) Forested wetlands Foe Freshwater wetlands Saline wetlands ed Semi-arid woodlands (Grassy subformation) es Dry sclerophyll forests (Shrub/grass subformation) Ee Semi-arid woodlands (Shrubby subformation) ae Dry sclerophyll forests (Shrubby subformation) oe Heathlands ee | Arid shrublands (Acacia subformation) Ez Arid shrublands (Chenopod subformation) Figure 1. Vegetation map of New South Wales (version 3.0) showing the distribution of 16 formations and subformations (see Table 1 and Keith 2004 for description) prior to intersection with extant mask. 2008). In total, the combined spatial dataset mapped the distribution of 833 geological units across New South Wales and was re-projected to the Australian Geodetic Datum 66 and Lamberts Conformal projection in a geographic information system. As for the vegetation data, the map legend was simplified by Proc. Linn. Soc. N.S.W., 132, 2011 assigning each lithological unit to one of 16 substrate types (Table 2). Substrate types were defined on the basis of attributes that were considered important influences on the supply of plant resources: mineral composition; weathering characteristics; texture and depth of derivative soils. The resulting map is shown in Fig. 2. ZEN GEODIVERSITY AND VEGETATION LATIONSHIPS BETW RI oinsodxo SOByIMS WINIsoUseU pue UINTO;RO YOG SUS SOSLICUUOS YOIYM “d}ILUOWWOP SopnyjoUl Os ‘sores BuosuNg ‘SoARD “W]< WINISOUSPUL:UINIO[vO YSIY YIM sjios Sulonpoid orydsourejout 2 duo}SOjOp UID] JeY-doajg uR]OUDL ‘oUO}SOJOp ‘oUOjsoWMIT = ATeOIdAy ~— Sue] Avjo pue she = ‘JuUdUOdUOD ROI]IS o[qeIARA YIM o}O]VO UT YOY —- Aaeyuoutpas “QUO]SOWNT] SuLIOy ROM a]YO1d uo Surpusdap JOAR] JSNOLINP s9Ry.ANsqns 10 QUO0Z snioydsoyd umissejod Jo suone.udsu09 (njIs ur) DOB]ANS “ULeIID) Je UO AypeoTdAT, ple ASN UL poqorsor A10,, wy|> SUIvOT MO] “WINTUTUUN]e “UO JO SUOTR.yUDDUOD YSI YMjOsoyy OLID}VT “UIR.LIO} JE] ‘spuryjayqe) pruuny oyeroduu) ul sojduuexe duos *au0Z Plie-1was dB10IH aye7T ‘oye 7] osuny| uIf< SUIYORI] 27 OUNISOI UOTZepUNUT ‘dd.1N0sS ({eIAn]e) S]USUUIpos JO soye] yerowoyds Apso ‘spoq oye] Jerowoyda io Arq = Ayeo1dAy_~—s Spures poyrunty ap sj]Ig uo Surpuodop ‘sjuoLyNU ysOU JO S[OAd] OJPIOPO|| Yposoy ouLysnoe 7] SeBHO| WINIUIUN]e pue UO. JO S[dAg| Apurs 0} SuioyyvoM ysiy ‘suoneo sjqvosueyoxe puv snsoydsoyd XLJPUL POUlRIS-OU —_ 70 SJOAd] MOT YIM J}IUTTORY 10 OYLIOpIS se Yyons juoudoyjaaap aj yoid esij]ig ‘sulseq Uo}IOJA[-d0USIR[D wy> ® Ul Sofonsed zj.1enb S[RIOUTUM YOLI-UINTUTUUN]e IO -UOMI JO XLQeUI B AreyUsUIpos [LOS poyruuy “ure119) yeY-doa}g a AoupASsouojspurs ssozjieng) — Ayjeard Ay pouleis-asiv0D, UI puNnog suIeIs zZjIenb osivOd JO souRpunqy AlejusUIpeg zjenb ys unisouseu ‘snioydsoyd ut yus1oyop yeyMouos spues[qe} Wroyjnos ske[o pur jnq “wuNIpos 29 wunisssejod “wunruTUNye Fo pue pue[sug MON U1a}SoM OU} wy]> SUILO] 0} SULIDYJLOM — SOA] YSIY YIM S[los Buronpoid (asvjoorseld 2 SOTUBS[OA uted} daajs AjjeoidAy, ~—- JO Saytoep ‘saqyruads ‘soytjoAyYy = AyeordAy PoulvIS-oULy oSe[IOY}IO ‘zj1eNb) sjeIOUTUT IS|O} JO S[OAd] YSIY snoous] dIS[94 unisouseu ‘snioydsoyd ul yus1oyop yeyMouos syyoujeq esog pur jsinyjeg Wy> oWOS SUIvO] yng “WInIpos 2p UINIsssej}od “wunTUTUIN]e Jo sdo}j][1y UO $10) YIM souoWOS ‘pur[suq MON ‘Soz]eRUo} Wy < Apues 0} SULIOYJVOM —-sTOADT -YSIY YIM s{Ios Sutonpoid (asepoorseyd a> SOAISNIQUI ‘ureia) Sunepnpun-dsa}g ‘so}oIpourss ‘soyiueiy = Ayyeord Ay PoUIvIS-dSIVOD 9Se[DOY}IO ‘zjyIeNb) s[eIOUILU SIS]TO} JO S[OAS| YSI snoous] dIS]Oy (jelang ulel19} jBIseOd = (ATeNyso JOATY 19j}UN}{) pueys] WI] < /ouLIeUut) S}USUUIpos JE] “UISIIO [RIAN a suey suvseiooy ‘sjeypnui yepry Ayyeord Ay SHES OPLIO]YS wWinipos Jo uOeUDDUOS Y SI} Yjosoy ouLenys| }svoo purjs] wf[> (ouTIeU) Jeou “UIR.LIO) SuUNR|NpuN-je[J IMOP Plo uo pajyorsos Alo, = Ay word Ay suo] Apues Q}eUOGIV WINIOTVO UI YorY ArejyUoUNIpag =, NIIULIBOTRD JOLNSIP dvd UI MSN SOJCUN]D PLIP-IWUAS UONPUWLO UdULIOOA\ ‘suIey;dues Ae]D AAS UT [lOsqns snooieoyRO ‘suOT}eO d[qeuosuRYyoxo (ueljoor) sulejdpues WY pur ‘uleL9) SuNnejnpun-jey 4 pue spjoyounp pues poy W][< iM weoy Apues oury pue snioydsoyd Jo s[aaa] o}e19pow-Mo7T yposoy (po) ueljooy sAvjo SuTyo[NUI -Jjos AAvoy sopnyoul susodap Arewioyend ‘sured [tos Os|y ‘pnut s1ues10 suoTeo d]qvosueYyoxo wnIAn|e yor|g pue sdwemsyoeq ‘suieyd ‘sureyd 9910 [AI IOAN PUuOUTY Sy Ppostjeoo] pue Kejo Sulos 9) snioydsoyd JO S[SA9] o}eIOpoUul-Yysry UyIM ({eIAnq]e) ule] dpooy *SOOAQ] “UIRLID] SuNe[Npun-je]y = “sureydpooy puryur a [eIsveog WI[< UjiIM QIIs 2 pues oul sAv[o puv spnut ‘spuvs our ‘s}]Is [RIAN] Yp[OsOY dATIOY SONSLIDIOBILYS IOIIO so]duiexq wdoq o1N}XoL, JUDJUOS [RIOUITA| ulsliQ, odd} ayeQSqns SITBAA INOS MAN] Ul SIdA] 97.QSQNS [VIISO[0IS JO SIISLIDJIVALYD “7 IIQUL, Proc. Linn. Soc. N.S.W., 132, 2011 10 D.A. KEITH AYSOI SOUINOWIOS “UTeLIO} yeP-doe}S$ sjisodop Arelploy, “UlelI9} Suye;npun-4ey 7 UleI19} ye] UO soyeuNITS plie-tues Jo Ajureur ‘sureyd [RIANT[e OAOCUT 2p SUIBO]S IOLIg UleI19} Je] 7 (sounp puelpeoy poyoied “3°9) uonisodaper URI[ODNe SUIOS jnq oUTIeW AT}sOUN ‘posijospod uajyo ‘sounp pue 24 auIjuodies 1eMopUeN] ‘so}UNp ‘soqopriiod ‘soytunuediag ule;doueg Ieqo_ sy} JO sodojs 1aMo] 29 SUISIVJ\Y “Sodoys 90} ‘saar0S sule[dpooy yseorapysea -ollenbory] oy} Uo sasis “ureyd AeY ‘sureyd pues “soyjouny] ‘spoq uleals yUopsoqjUVy UOHeULIO; uoyieddays ‘urejd eulioAny sseul pues suledpurs ey ‘[eyseoo-Ieoy] solopuning “sureyd pues jejseo_d so] youd [Ios past19je] dosp sulonpoid saumjouos JOWIOF oy} ‘daoys Aj]euotseos0 ‘UIeLO} SUL]NpUN-jey USO UIe.119} yey-daa1s SOLUBOTOA SUOSULIOD ‘sureyd Joodioary] “O1gqqes ‘oylIa[0p “ayeseg ‘Aevajory|-Aosdy “S|SIYOs ‘soypAyd ‘oyowmAois “So[eYs “SOUO}SpNU “SSUO}S}IIS W]> suieoy Aejo pue shvjo 0} AjyeordA} SurrayjeoM pouress-ouly UWI[< S[oAvis 79 suo] Apues wiv< AypeardAy S}[IS 27 spurs Wy < sueo] Ava ws =I Aypeord A} spurs uimissejod pue snioydsoyd UI SOLNUSTOYOp pue ‘fox pue WINTUTOIYS “UOT JO S[oAo] YSIY ‘sored WnIOTeo :wUNIsoUseU W]< swieoy Avjo pure sdvjo 0} Zuronpord (a1 01g ‘ojoqrydure ‘outArjo ‘ouoxo1Ad AyjeoidAy SuLoyjyeom pouleis-ouly Wf< Asour wieoy 2 AelD ysiy AIOA YIM S[IOs Suronpoad “eos Jo sjaAsy] — o1yd1ourejoUI MO] PUL S][PIOUIU OUI JO sjoAo] Ysiy AIaA, ‘snoous] (Teranqye suoled o[qeesueyoxe /[etAnyyoo) pue snroydsoyd Jo sjaag] a}e1opoul-MoT Yposoy (jeranyye) SJUDLINU JSOUW JO S[OAS] MO] 0} BJeIOPO/[ YMposay suones o]qeosueyoxo ([erAnqy) pue snioydsoyd fo s[ad] oye1OpoOUl-MoT YMposoy Aeids yes Aq psouonypur o1oyM (ourlreun) ydooxo ‘s}USLINU [PIOUTUN SOUT UI MO] AIOA-MOT UMposoy UOI “WnIsouse un ‘snioydsoyd fo sjaao] Ysty ATOALIOI YIM STIOS ‘Qpud[qUIOY) S[eIOUTU SLU JO SOAS] YS] snoous] suones sjqeosueyoxo 29 sniodsoyd Jo sjoag] YsIy-o}e1OPOU! YIM S[los Suronpoid ‘oy ‘oyLoyeo “zyrenb Jo sajonred WIS YIM poxtur (eI0TYS “oyuTfoRy “onty! — orydrourejout “OUT | [IIOWJUOU “3°O) SjeIOUTU Ae[D JO oANyXIP\, 2 AreyUOUIPES sorydrourejour 3 SNOSUSI oyeMmeny [OAvIS 29 pues [erAnyye \ /[etAnyyoo [enpisoy Spurs |eIAnT[e [enpisoy Avjo [eranyye AMMEN suie[dpues (oy) SnoddI{IS ouLey SOAISNIJUL 2 SOLUBITOA ogre AreyusuIpas Zy1enb MOT 11 Proc. Linn. Soc. N.S.W., 132, 2011 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION 0 50100 200 300 400 Kilometres Geological Substrates bad aeolian (red) sandplains BE ed limestone me estuarine sediments me low quartz sedimentary HD felsic intrusives WE matic volcanics & intrusives hal felsic volcanics esd residual alluvial clay HI floodplain alluvium ___ residual alluvial sands Be! 24 high quartz sedimentary ees residual alluvial/colluvial sand & gravel WM lacustrine sediments siliceous (white) sandplains beers laterite eee ultramafic igneous & metamorphics Figure 2. Geological map of New South Wales showing the distribution of 16 substrate types (see Table 2 for description). 12 Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH Climate surfaces Spatial grids for monthly precipitation, temperature and radiation parameters were generated by fitting spline functions (Hutchinson 1991) to weather station data across Australia. Using ANUCLIM v6.1 (http://fennerschool.anu.edu.au/ publications/software/anuclim.php#overview), the weather station data were aggregated to monthly averages for a 30-year period (1975-2005) and interpolated across the continent on a 9-second latitudinal-longitudinal grid (approx. 250 m). The weekly climate averages were used to calculate the following parameters: MeanTemp- Mean temperature across all weeks of the year; MinTmp- mean of lowest weekly minimum temperature; MaxTmp- mean of highest weekly maximum temperature; DiRngImp- mean of weekly diurnal temperature ranges; AnnRain- Mean annual rainfall; SummRain — Mean rainfall of December-February; RainDryMth- Mean rainfall of the driest month; MIdry- Mean moisture index of the driest month. Moisture index is calculated from weekly rainfall, evaporation and soil moisture storage (see http://fennerschool.anu.edu. _ -au/publications/software/anuclim/doc/params.html). Spatial data for these parameters were re-projected to the Australian Geodetic Datum 66 and Lamberts Conformal projection in a geographic information system. Data analyses To examine the fidelity between vegetation types and geological substrates, 500 randomly located points were generated in a GIS to sample each of the 100 mapped vegetation classes. Subsequently, these were randomly sub-sampled to obtain 1000 points in each of the 16 vegetation formations and subformations. The points sampling vegetation formations and classes were intersected with the geology map (i.e. vegetation cross-tabulated with geology) to estimate their frequencies of occurrence on each substrate type. Similarly, 1000 randomly located points were generated to sample each of the 16 mapped geological substrates, and these were intersected with the vegetation map to estimate the diversity of vegetation formations and classes represented on each substrate type. Calculations of the number of points represented in the cross- tabulations were based on 90" percentile of points to reduce the effect of boundary errors in mapping that may cause spurious occurrences of vegetation types on particular substrates. Thus, the number of substrate types per vegetation unit was taken as the minimum number of substrates accounting for 90% of points Proc. Linn. Soc. N.S.W., 132, 2011 within that unit. Pearson correlation coefficients were calculated to assess the association between the area of vegetation classes and formations and the number of substrates that they occupy. The variation in species composition among vegetation classes attributable to geological substrate and climate was evaluated using ordinations and partial variance analyses (Leps & Smilauer 2003). This required construction of three data matrices characterising the species composition, geological substrates and climatic habitat of vegetation classes. A presence/absence species matrix (99 classes x 1625 species) was constructed from the floristic descriptions (vascular flora) of vegetation classes in Keith (2004), which had been compiled from frequently mentioned species and identified dominant Species in the descriptions of source map units. A substrate matrix (99 classes x 16 geological substrates) was constructed from the relative representation of substrates types within each mapped vegetation class, estimated from frequencies of 500 random points per class, as described above. A climate matrix (99 classes x climate variables) was constructed from mean climate parameters across the same samples of 500 random points, which were intersected with each of the nine climate surfaces. Unfortunately, climate surfaces were unavailable for offshore areas, so four vegetation classes (Oceanic Rainforests, Oceanic Cloud Forests, Coastal Headland Heaths and Maritime Grasslands) were omitted from the analysis. Hence the species matrix was reduced to 95 classes x 1500 species. To determine whether species responses best fitted a linear or unimodal response to environmental gradients, redundancy analysis and canonical correspondence analyses were each carried out on the species matrix constrained by the combined matrices for geological substrates and climate parameters. The first four Eigen vectors of each analysis accounted for a similar proportion of floristic-environmental relationships (25%), although the redundancy analysis accounted for slightly more floristic variation than the canonical correspondence analysis (7.8% cf. 7.2%). A linear response model was therefore assumed for subsequent analyses. An unconstrained principal components analysis was first carried out on the species matrix. This allowed display of floristic relationships and examination of environmental relationships using indirect gradient analysis to fit vectors representing each variable in the substrate and climate matrices to the floristic ordination. Two partial redundancy analyses were then 13 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION carried out to quantify the proportion of floristic variation uniquely attributable substrate types and climatic variables. These were done as constrained ordinations with substrate and climate defined as the environmental and covariable matrices, respectively, and then vice versa for the second analysis (ter Braak & Smilauer 1999). A third partial redundancy analysis was then carried out to determine the combined (union) proportion of floristic variation attributable to either substrate or climate with both sets of variables combined within a single environmental matrix. Finally, the (intersection) proportion of variation attributable to both substrate and climate combined was calculated by subtracting the sums of all canonical eigen values from the first two partial redundancy analyses from the sum of canonical eigen valyes in the third partial redundancy analysis (Leps & Smilauer 2003). All redundancy analyses were carried out in CANOCO for windows v4.02 (ter Braak & Smilauer 1999). RESULTS Fidelity of vegetation types to geological substrates While none of the vegetation formations was restricted to a single geological unit, there was a strong non-random relationship between vegetation 100 “% of formation extent ona single substrate type 2 Number of substrates per vegetation formation 4 formation and geological substrate (y? = 21562, P<<0.001, df = 225). Approximately 27-61% of the extent of each formation occurred on one substrate. More than 90% of the distribution of ten of the 16 formations was restricted to five or less geological substrates and all but one formation was restricted to seven or fewer substrates (Fig. 3). The number of substrates occupied by a vegetation formation was unrelated to the extent of its distribution (R=-0.078, P>0.5, df=15), indicating that vegetation formations were not restricted to a small number of substrates simply by virtue of small distributions. Furthermore, geological substrates supported a limited range of vegetation formations. Four or fewer vegetation formations accounted for more than 90% of the area covered by eleven of the 16 substrate types and none of the substrate types supported more than seven formations. The number of formations per substrate type was unrelated to substrate area (R=0.43, P~0.1, df=15). At the level of vegetation class, the association with geological substrates was even more strongly expressed. Seven vegetation classes were essentially restricted to a single substrate type, while two-thirds of the 99 classes had at least 90% of their distribution restricted to three or fewer substrates (Fig. 4). As for formations, the number of substrates occupied by a vegetation class was unrelated to the extent of its distribution (R=0.027, P>0.5, df=98). 6 8 10 Figure 3. Fidelity of 16 vegetation formations to 16 geological substrate types. 14 Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH 6 8 10 Number of substrates per vegetation class Figure 4. Fidelity of 99 vegetation classes to 16 geological substrates. rs 100 = o wo — SSS = 80 =o 7 @. yoo oom G2 o Zz 40 i=] 2 2 - = Oo w = 0 0 & Rainforests, both Wet Sclerophyll Forest subformations and Grassy Woodlands were strongly associated with low-quartz sediments and metasediments (Appendix 1). Rainforests also occurred frequently on felsic volcanics, while the wet sclerophyll forests and grassy woodlands were more strongly associated with felsic intrusives. It is likely that the rainforests, grassy woodlands and grasslands were also well represented on mafic volcanics, but much of this substrate has been cleared of its native vegetation. The Heathlands and shrubby subformation of Dry Sclerophyll Forests were strongly associated with high-quartz sediments and siliceous (white) sands of marine origin, but also had significant representation on low-quartz sediments. In contrast, the shrub/grass subformation of Dry Sclerophyll Forests was primarily associated with low-quartz sediments and felsic intrusives (Appendix 1). The Alpine Complex occurred mainly on felsic intrusives and low-quartz sediments. All three wetland formations and the Grasslands were strongly associated with active fluvial alluvium, with lower frequencies of occurrence across a range of other substrates. The shrubby Semi-arid Woodlands and Arid (acacia) Shrublands were strongly associated with aeolian (red) sands, while the grassy Semi-arid Woodlands occurred primarily on floodplain alluvium and residual alluvial clays and Arid (chenopod) Shrublands were on aeolian sands and residual clays (Appendix 1). The vegetation classes that were essentially restricted (>90% of occurrence) to one substrate type Proc. Linn. Soc. N.S.W., 132, 2011 included a range of Rainforests, Dry Sclerophyll Forests, Heathlands and Semi-arid Woodlands (Appendix 2). The geological substrates that supported the broadest ranges of vegetation formations include low- and high-quartz sedimentaries, felsic intrusives and floodplain alluvium (Appendix 1). Relative influence of geology and climate on vegetation The Principal Components ordination showed that vegetation classes within the same formation generally clustered together (Fig. 5a). This suggests considerable floristic affinities within formations, even though classes were grouped together within formations on the basis of structural and functional resemblance, rather than compositional resemblance. Indirect gradient analysis showed that geological substrates and climatic parameters account for a diverse array of compositional gradients within vegetation of the region (Fig. 5b). Individual climatic parameters appeared to exert a stronger influence on vegetation, as their vectors were generally longer than those representing individual geological substrates, indicating stronger correlations with species composition. However, geological substrates appeared to exert a more diverse range of influences, as their vectors spanned a greater range of directions than those representing climate parameters (Fig. 5b). Partial variance analysis showed that, in combination, the full set of geological substrates accounted for a greater proportion of variation in species composition than the climate variables (Fig. 6). 15 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION «RF « VWWSFs oVWYSFg - GV x GL a DSFg 2DSFS + HL x AC - Fry 2 Fovil e SV 2 SAVYG e@ SAVYS @ASc oASa — % SS Ss ae?” S F %, 2 ae rap rales Re se - —Vean tmp Ran? wih A cp SootSand Salt 8) XS, Mey, Bery Sr o eos eal %, vy po QW oF Zs. & 2 = 2 a Figure 5. (a) Scatter plot of unconstrained Principal Components Analysis of 99 vegetation classes grouped by formations: RF- Rainforests, WSFs- Wet Sclerophyll Forests (shrubby subformation), WSFg- Wet Sclerophyll Forests (grassy subformation), GW- Grassy Woodlands, GL- Grasslands, DSFg- Dry Sclerophyll Forests (shrub/grass subformation), DSFs- Dry Sclerophyll Forests (shrubby subformation), HL- Heathlands, AC- Alpine Complex, FrW- Freshwater Wetlands, FoW- Forested Wetlands, SL- Saline Wetlands, SAWg- Semi-arid Woodlands (grassy subformation), SAWs- Semi-arid Woodlands (shrubby subformation), Arid Shrublands (chenopod subformation), Arid Shrublands (acacia subformation). (b) Plot of vectors representing 16 geological substrates (thin black lines) and 9 climate parameters (thick grey lines) fitted to the Principal Components ordination. The substrate types are: ResSand- Residual Alluvial Sand, Colluvial- Colluvial/alluvial sand and loam, ResClay- Residual alluvial clay, ActAlluy- Active alluvium, Lacust- Lacustrine sediments, EstSeds- Estuarine sediments, MarSand- Marine sands, Hqseds- High-quartz sedimentary rocks, Lqseds- Low-quartz sedimentary & metamorphic rocks, FelsInt- Felsic intrusives, FelsVolc- Felsic volcanics, UltMaf- Ultramafic volcanic and metamorphic rocks, Limestn- Limestone, MafVolc- Mafic volcanics. The climate parameters are: DiRngTmp- Diurnal range of temperature, MaxTmp- Mean temperature of the warmest month, MeanTemp- Mean annual temperature, MIdry- Mean moisture index (see text) of the driest quarter, MinTmp- Mean temperature of the coldest month, AnnRain- Mean annual rainfall, SummRain — Mean rainfall of December-Febru- ary, RainDryMth- Mean rainfall of the driest month. Note the vector plot (5b) is enlarged by a factor of 2 relative to the scatter plot (5a). 16 Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH Figure 6. Venn diagram showing portions of variation in floristic composition of vegeta- tion classes attributable to substrate only (S), climate alone (C), both substrate and climate (SC) and unexplained variation (U), as determined by partial redundancy analysis. Geological substrates and climate accounted for largely independent components of variation in species composition, as only 4.4% of total floristic variation was correlated with both geology and climate in combination. Together, geology, climate and their overlapping component accounted for just over one-third of total floristic variation, leaving two- thirds unexplained (Fig. 6). DISCUSSION The influence of geodiversity on vegetation Vegetation exhibited strong relationships with geodiversity at both class- and formation level across 80 million hectares of south-eastern Australia. Almost one-fifth of floristic variation across this large temperate region was uniquely attributable to geological substrates, independent of climatic variables. Each vegetation formation and class showed strong fidelity to a small range of geological substrates, with some classes restricted to a single substrate type. Stronger fidelity at the class level, relative to vegetation formations, indicates that relationships between vegetation and geodiversity are scale-dependent. At finer levels of vegetation Proc. Linn. Soc. N.S.W., 132, 2011 classification than class, a still greater proportion of plant assemblages are restricted to a single type of substrate (e.g. Tozer et al. 2010). Indirect gradient analysis showed that species composition of vegetation was more strongly correlated with individual climate parameters, notably rainfall, than any single substrate type. However, the compositional trends associated with substrates encompassed a broader array of gradients than those associated with climate parameters. As a consequence, partial variance analysis showed that the substrate types collectively accounted for more variation than a set of parameters encompassing the means and extremes of climatic moisture, temperature and patterns in their seasonality. The overlapping component of floristic variation attributable to both geodiversity and climate was remarkably small. Amongtheclimatevariables, floristicrelationships with the three rainfall parameters were positively correlated with one another and negatively correlated with vectors representing maximum temperature, diurnal range and seasonality. This major gradient was associated with the transition from forested vegetation classes to semi-arid woodlands and arid shrublands. Minimum temperature and moisture index of the driest month (which incorporates evapo-transpiration 17 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION as well as precipitation) displayed somewhat different floristic trends. A strong contrast was evident between substrates that produce (high-quartz sediments, marine (white) sands) and those that produce more fertile soils (low-quartz sediments, felsic intrusives and volcanics). The former were strongly associated with vegetation types dominated by sclerophyllous shrubs (as understorey or canopy species), while the latter were associated with vegetation types with abundant mesophyllous shrubs and/or grasses. The mafic volcanic substrates generally define the upper limit of this soil fertility gradient, while vectors representing substrates with extreme levels of some mineral elements (ultramafics, limestone) are intermediate between those of mafic and felsic substrates. A similar but more subtle distinction is evident between vegetation classes found within dry-climate regions. Shrubby semi-arid woodlands and arid (acacia) shrublands are associated primarily with impoverished aeolian (red) sands and residual alluvial sands, clays and colluvium, while grassy semi-arid woodlands and arid (chenopod) shrublands are more common on active alluvium and residual clays. Estuarine and lacustrine sediments are uniquely associated with various types of wetlands, which are also associated with active fluvial alluvium. impoverished — soils Support for biogeographic landscape theories The patterns described above are consistent with early comparative work between the flora of low- and high-quartz substrates in the Sydney region (Beadle 1953, 1966) and with soil-vegetation relationships inferred from early survey work in western New South Wales (Beadle 1948). This work highlighted the association between sclerophylly and soil nutrients, notably phosphorus, which are more abundant in clay minerals derived from mafic substrates than felsic substrates and least abundant in quartz-rich substrates (Table 2). The observed vegetation-substrate patterns generally support Hopper’s (2009) characterisation of two general landscape types: Young Often Disturbed Fertile Landscapes (YODFELs) and Old Climatically Buffered Infertile Landscapes (OCBILs). ‘Young’ and ‘old’ in Hopper’s sense refer to age of landscape, rather than underlying geology. Hence YODFELs are characterised by relatively fertile soils whose nutrient capital has not been greatly depleted by leaching and which may undergo frequent disturbance related to fluvial or maritime events or mass movement. Their flora is dominated by recently evolved species with long-distance dispersal capabilities, propensity for 18 colonisation, extensive distributions, generalist nutritional and reproductive biology, and tolerance of disturbance (Hopper 2009). The YODFEL profile fits many species of the grassy vegetation formations and subformations, which occur on the more fertile substrates (e.g. low-quartz sediments, volcanics, active alluvium). It also generally fits a large portion of the flora that characterises the three wetland formations, which may generally be viewed as occupying resource-rich sinks within regional landscapes (Keith 2004). In contrast, OCBILs are characterised by a diversity of ancestral species lineages with limited dispersal and colonisation capability, often with restricted distributions, specialised nutritional and reproductive biology, prominent sclerophylly and limited resilience to physical disturbance. Additional species traits associated with the sclerophyll syndrome were described in mechanistic detail and for a broader range of biota by Orions & Milweski (2007) in their ‘‘Nutrient-Poverty/Intense-Fire Theory”. The OCBIL profile describes many of the sclerophyll plant species that characterise substrates associated with impoverished soils (e.g. high-quartz sediments, leached marine sands, aeolian sands). Both landscape types appear to extend throughout the humid — arid climatic gradient of the region. It is noteworthy that much of rainforest flora does not readily fit either profile. Many of the taxa occupy climatically buffered environments and belong to ancient lineages that generally lack recent radiation and have suffered numerous extinctions (Crisp et al. 2004). Yet their habitats are not the most nutrient-impoverished nor very ancient landscapes and many of the taxa are widely dispersed with large distributions, some are ready colonisers. Axiomatic to both Nutrient-Poverty/Intense-Fire and OCBIL theories is the proposition that plants growing on nutrient-deficient soils with periodically adequate moisture, can synthesize ‘excessive’ carbohydrates, which are deployed to produce well- defended foliage, large quantities of lignified tissues and readily digestible exudates (Orions & Milewski 2007, Hopper 2009). The nutritional properties of geological substrates therefore define a fundamental basis for evolution of Australian biota and retain a distinctive signature on the present-day distribution of vegetation formations and assemblages in the region of south-eastern Australia examined here. Given their strong influence on contemporary vegetation patterns, geological substrates which, with few exceptions, are essentially fixed landscape features over millenial time scales, appear to impose significant constraints Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH on vegetation response to climate change, especially in landscapes with OCBIL characteristics. Approximately two-thirds of the floristic variation remained unexplained in the direct gradient analysis. Part of this unexplained variation may include unrepresented influences of soils and climate. For example, substrate types were defined very broadly and often encompass considerable heterogeneity, not only in the complexes of rocks juxtaposed within them, but in the mineral composition and texture and structure of soils produced across catenary sequences of the landscape. The movement and availability of water across the landscape is also an important source of variation that is not fully represented by the climatic variables included in the current analysis. This essential resource almost certainly accounts for some of the unexplained floristic variation, particularly in the wetland component of the biota. Fire regimes are also likely to account for a fraction of the unexplained variation, as a lack of suitable spatial data precluded any consideration of them in the analysis. Fire regimes have been identified as driving evolutionary forces in Australia and other continents (Bond 2005, Bowman et al. 2009). They are an important component of Nutrient-Poverty/ Intense-Fire theory, as rapid accumulation of nutrient- poor biomass, a result of low rates of herbivory, provides fuel for intense fire, which in turn promotes nutrient poverty through volatilisation (Orions & Milewski 2007). Any remaining variation in floristic composition of south-eastern Australia is mostly attributable to sampling error and inherent spatial autocorrelation, as time lags in vegetation dynamics and limited dispersal processes impose an inherently clustered spatial structure on the composition of biota in the landscape. Map-based approach to ecological analysis The map-based approach employed in this study has both strengths and limitations. A major advantage is that it permits a balanced stratified random sampling design across the entire study area. This overcomes a significant constraint for analyses based on field samples over such a large region — the available data are inevitably skewed and non- randomly distributed across the landscape to varying degrees. A complementary analysis based on field samples may nonetheless be profitable, as it permits a more direct location-based exploration of vegetation- environment relationships, and hence exploration of finer-scale patterns than can be represented on sub- continental maps. A potential limitation of the map-based approach is that imprecision in the boundaries of both maps Proc. Linn. Soc. N.S.W., 132, 2011 may have resulted in some combinations of vegetation and substrate types that do not occur in nature, as well as a margin of error in estimated frequencies of association. Non-concurrence of soil, soil parent material and bedrock could occur, for example, where there is significant lateral movement of sediment downslope from its origin. This promotes a tendency for the fidelity of vegetation types to substrate types to be under-estimated (i.e. vegetation types are more restricted to vegetation types than the data indicate). To offset such effects, frequencies in Figs. 3 and 4 were based on the 90" percentile of sampled points, although it is uncertain whether this adjustment adequately compensated spatial errors in the absence of field validation data. A second limitation of the map- based approach is that it does not allow relationships between floristics and environmental data to be explored directly. This was because the species and substrate matrices were based on descriptive data averaged across the mapped range of each unit, rather than location-specific estimates. Thirdly, depending on the methods employed to generate source maps of for corresponding areas, the spatial data for vegetation and geology may not be independent throughout the mapped area. For example, in some cases geological boundaries may have been used as proxies for vegetation mapping and conversely remote sensing of vegetation may have been used to identify geological boundaries. Such non-independence may inflate map- based correlation between vegetation and geology. However, such effects are mitigated by the use of multi-criteria in remote sensing and modelling, only some of which will be non-independent proxies, as well as varying levels of field sampling to directly verify mapped units. Independent reclassification of the vegetation and geological maps further reduced any non-independence. While these methodological issues limited the resolution of relationships that could be examined, the analytical methods employed were sufficiently sensitive to detect major influences of geological substrates on vegetation that, collectively, appeared to be stronger than, and largely independent of climatic influences. ACKNOWLEDGEMENTS I thank the organisers of the Geodiversity Symposium, especially Ian Percival who encouraged me to present a paper on relationships between vegetation and geodiversity. Chris Simpson assisted with compiling the spatial data for analysis. Mike Hutchinson and Tingbao Xu prepared the climate surfaces under a collaborative research project funded by the Australian Research Council (LP0989537). 19 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION REFERENCES Andrews, E.C. (1916). The geological history of the Australian flowering plants. American Journal of Science Series 4 42:171-232. Beadle, N.C.W. (1948). The vegetation and pastures of western New South Wales. Department of Conservation of New South Wales, Sydney. Beadle, N.C.W. (1953). The edaphic factor in plant ecology with a special note on soil phosphates. Ecology 34:426-428. Beadle, N.C.W. (1966). Soil phosphate and its role in moulding segments of the Australian flora and vegetation, with special reference to xeromorphy and sclerophylly. Ecology 47: 992-1007. Bowman, D.M.J.S, Balch, J.K., Artaxo, P., Bond, W.J., Carlson, J.M., Cochrane, M.A., D’Antonio, C.M., DeFries, R.S., Doyle, J.C., Harrison, S.P., Johnston, F.H., Keeley, J.E., Krawchuk, M.A., Kull, C.A., Marston, J.B., Moritz, M.A., Prentice, I.C., Roos, C.1., Scott, A.C., Swetnam, T.W., van der Werf, G.R., Pyne, S.J. (2009). Fire in the Earth System. Science 324: 481-484. Bond, W.J. (2005). Large parts of the world are brown or black: A different view on the ‘Green World’ hypothesis. Journal of Vegetation Science 16: 261- 266. Christian, C.S. (1952). Regional land surveys. Journal of the Australian Institute of Agricultural Science 18:140-146. Crisp, M.D., Cook, L. and Steane, D. (2004). Radiation of the Australian flora: what can comparisons of molecular phylogenies across multiple taxa tell us about the evolution of diversity in present-day communities? Philosophical Transactions of the Royal Society London B 359:551—1571 Crisp, M.D., Arroyo, M.T.K., Cook, L.G., Gandolfo, M.A., Jordan, G,J., McGlone, M.S., Weston, P.H., Westoby, M., Wilf, P., Linder, H.P. (2009). Phylogenetic biome conservatism on a global scale. Nature 458:754—756 Crocker, R.L. (1944). Soil and vegetation relationships in the lower south-east of South Australia, a study in ecology. Transactions of the Royal Society of South Australia 68:144-172. Diels, L. (1906). The plant life of Western Australia south of the tropics. Wilhelm Engelmann, Leipzig. Fraser, L. and Vickery, J.W. (1939). The ecology of the upper Williams River and Barrington Tops districts. Ill. The eucalypt forests, and general discussion. Proceedings of the Linnean Society of NSW 64: 1-33. Gray, M. (2004). Geodiversity: Valuing and Conserving Abiotic Nature. John Wiley & Sons Ltd, Chichester. Hannon, N.J. (1956). The status of nitrogen in the Hawkesbury sandstone soils and their plant communities in the Sydney district:.I. The significance and level of nitrogen. Proceedings of the Linnean Society of New South Wales. 81:199-143. Hill, R.S., Truswell, E.M., McLoughlin, S. and Dettman, M.E. (1999). The evolution of the Australian flora: 20 fossil evidence. In ‘Flora of Australia’, vol. 1, second edition. Introduction (ed. A. E. Orchard), pp. 251- 320. (CSIRO, Melbourne). Hopper, S.D. (2009). OCBIL theory: towards an integrated understanding of the evolution, ecology and conservation of biodiversity on old, climatically buffered, infertile landscapes. Plant and Soil 322: 49-86. Hutchinson, M.F. (1991) The Application of thin plate smoothing splines to continent-wide data assimilation. In Data Assimilation Systems, J.D.Jasper (ed), Bureau of Meteorology Res. Rep. No. 27, Bureau of Meteorology, Melbourne, pp. 104- il). Johnson, L.A.S. and Briggs, B.G. (1975). On the Proteaceae: the evolution and classification of a southern family. Botanical Journal of the Linnean Society 70: 83-182. Keeley, J.E. and Rundel, P.W. (2005). Fire and the Miocene expansion of C-4 grasslands. Ecology Letters 8: 693-690. Keith, D.A. (2004). Ocean shores to desert dunes: the native vegetation of New South Wales and the ACT. NSW Department of Environment and Conservation, Sydney. Keith, D.A. and Sanders, J.M. (1990). Vegetation of the Eden region, South-eastern Australia: Species composition, diversity and structure. Journal of Vegetation Science 1: 203-232. Keith, D.A. and Simpson, C.C. (2008). A protocol for assessment and integration of vegetation maps, with an application to spatial data sets from south-eastern Australia. Austral Ecology 33: 761-774. Lambers, H., Brundett, M.C., Raven, P. A. and Hopper, S.D. (2010). Plant mineral nutrition in ancient landscapes: high plant species diversity on infertile soils is linked to functional diversity for nutritional strategies. Plant and Soil 334: 11-31. Lamont, B. (1982). Mechanisms for enhancing nutrient uptake in plants, with particular reference to mediterranean South Africa and Western Australia. Botanical Review 48:597-689 Leps, J. and Smilauer, P. (2003). Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge. McLuckie, J. and Petrie, A.H.K. (1927). An ecological study of the flora of Mount Wilson. Part iv. Habitat factors and plant response. Proceedings of the Linnean Society of NSW 52: 161-184. Orians G.H. and Milewski A.V. (2007). Ecology of Australia: the effects of nutrient-poor soils and intense fires. Biological Reviews 82: 393-423 Pidgeon, I.M. (1937). The ecology of the central coastal area of New South Wales. I. The environment and general features of the vegetation. Proceedings of the Linnean Society of NSW 62: 315-340. Shane, M. and Lambers, H. (2005). Cluster roots: a curiosity in context. Plant and Soil 274:101—125. Stewart, A.J., Sweet, I.P., Needham, R.S., Raymond, O.L., Whitaker, A.J., Liu, S.F., Phillips, D., Retter, Proc. Linn. Soc. N.S.W., 132, 2011 D.A. KEITH A.J., Connolly, D.P. and Stewart, G. (2008). Surface geology of Australia 1:1,000,000 scale [Digital Dataset]. Geoscience Australia, Canberra. http:// WWW.ga.g0v,au. ter Braak, C.J.F. and Smilauer, P. (1999). CANOCO for Windows version 4.02. Centre for Biometry Wageningen, CPRO-DLO, Wageningen, The Netherlands. Tozer, M.G., Turner, K., Keith, D.A., Tindall, D., Pennay, C., Simpson, C., MacKenzie, B., Beukers P. and Cox, S. (2010). Native vegetation of southeast NSW: a revised classification and map for the coast and eastern tablelands. Cunninghamia 11:359-406. Troedson, A.L. and Hashimoto, T.R. (2008). Coastal Quaternary geology — north and south coast of New South Wales. Geological Survey of New South Wales Bulletin No. 34. NSW Department of Primary Industries, Sydney. Webb, L.J. (1954). Aluminium accumulation in the Australian—New Guinea flora. Australian Journal of Botany 2:179-196 Proc. Linn. Soc. N.S.W., 132, 2011 J \\ ODIVERSITY AND VEGETATION EN GE =% | LATIONSHIPS BETWE 4 4 RI Appendix 1. Percentage of 1000 random points in each vegetation formation on each geological substrate type o = @ o 8 ¢ =} iss) Or au n i?) n n = 2 iss} 3 @ Q a = th Oo Dw = ee one es R. z. Se 2 8 oo ee | Cee we og © BERS BeoG8 ie Bo wee ge 8 ee sea ee eee is Substrate type: BB Be ER 5 Sa Sn a2 34 a cite Bigs fs, fees Be UEe 698 aS =) ee het = iy lis gb iS) 3 p Qiao E& GE BS Bee gas Se Ce ye G8) PEG ica oo CEG gee te ee eee * 2 oO 3 5. BS e & Be Vegetation Formation: i Alpine complex 0 0 0 59 3 ] 0 0 20 16 0 0 0 0 0 Arid shrublands 56 0 0 0 0 14 0 2 0 1] 0 0 0 16 0 0 (Acacia subformation) Arid shrublands 33 0 0 0 0 3} 0 5 0 11 0 24 0 14 0 0 (Chenopod subformation) Dry sclerophyll forests (Shrub/grass 5 0 0 22 6 2 10 0 0 47 3 0 0 4 0 0 subformation) Dry sclerophyll forests 0 0 0 11 5 5 36 0 0 28 3 0 0 2 9 0 (Shrubby subformation) Forested wetlands l 0 0 6 3 39 5) 12 0 21 7 ] 0 2 4 0 Freshwater wetlands 9 0 0 I] ] 33 11 a 0 8 ) 6 0 2 8 0 Grasslands 10 0 0 4 i 32 1 0 6 10 19 0 1 15 0 Grassy woodlands | 0 0 24 y] 5 6 0 38 12 2 ] 2 0 0 Heathlands 0 0 0 10 5 5 27 0 0 24 3 0 0 5 21 0 Rainforests 0 4 0 4 20 2 il 0 0 43 15 0 0 0 4 0 Saline wetlands 14 0 27 0 0 4| 6 12 0 5 0 0 0 2 16 0 Semi-arid woodlands 13 0 0 0 0 4] 0 ] 0 9 2 16 4 14 0 0 (Grassy subformation) Semi-arid woodlands 50 0 0 | 1 13 1 ] 0 13 0 5) 5 13 0 0 (Shrubby subformation) Wet sclerophyll forests 0 0 0 24 5 l 15 0 0 44 10 0 0 0 0 0 (Grassy subformation) Wet sclerophyll forests 0 0 0 20 5 | 8 0 0 61 5 0 0 0 0 0 (Shrubby subformation) —_—— ee: _O_ _ eee — —— Proc. Linn. Soc. N.S.W., 132, 2011 2 D.A. KEITH Appendix 2. Percentage of 500 random points in each vegetation class on each geological sub- strate type (see Keith 2004 for description of classes). = = 5 eee 60 Mp Ree [on S feb) z Subtropical Rainforests OO 0 5 2 Northern Warm Temperate Rainforests 0 0 0 6 10 CookTemperate Rainforests 0 O 0 9 1 Dry Rainforests ® @ 0 4 7 Littoral Rainforests 0 0 0 0 North Coast Wet Sclerophyll Forests 0 0 0 4 3 South Coast Wet Sclerophyll Forests 0 0 0 13 5 Northern Escarpment Wet Sclerophyll 0 0 0 9 9 Forests Southern Escarpment Wet Sclerophyll 0 0 0 56 3 Forests Northern Tableland Wet Sclerophyll 0 0 0 16 5 Forests Southern Tableland Wet Sclerophyll 0 0 0 33 D Forests Sydney Coastal Dry Sclerophyll 0 0 0 0 0 Forests _ Sydney Hinterland Dry Sclerophyll 0 0 0 0 0 Forests Sydney Montane Dry Sclerophyll 0 0 0 0 0 Forests Coastal Dune Dry Sclerophyll Forests 0 0 0 0 1 North Coast Dry Sclerophyll Forests 0 0 0 1 i Northern Hinterland Wet Sclerophyll 0 0 0 3 > Forests South Coast Sands Dry Sclerophyll 0 0 0 0 1 Forests Southern Lowland Wet Sclerophyll 0 0 0 8 > Forests Northern Escarpment Dry Sclerophyll 0 0 0 8? 7 Forests South East Dry Sclerophyll Forests 0 0 0 7) 8 Northern Tableland Dry Sclerophyll 0 0 0 36 71 Forests Southern Tableland Dry Sclerophyll 0 0 0 16 16 Forests Western Slopes Dry Sclerophyll 1 0 0 4 6 Forests Pilliga Outwash Dry Sclerophyll 48 0 0 0 0 Forests Wallum Sand Heaths 0 0 0 Sydney Coastal Heaths 0 O 0 Northern Montane Heaths 0 O 0 69 16 Sydney Montane Heaths @ © 0 0 0 Proc. Linn. Soc. N.S.W., 132, 2011 WINIANTe ureydpooy. SoS OS oS WU 19 Areyuourpes zyrenb ysiy is) 24 22 38 10 S}USUIIPaS oUTISNOv] Sa oo eo 2 8S So © Oo Oo So ©S& OYL19}L] S eo oF oOo Ee fF OS © S eo eo 2 © 90} SOUT] Sa eo ooo oO So 2S Se Seo 2 Se © Areyuouttpes zj1enb Moy] SOAISNLIQUI 29 SOIUBOTOA OYRUt 33 14 25 oS Xe S&S oS WS Avjo [eIAnqye [enpisoer Sa So eo oe 282 Oo SC OS So Eo So So ©S& spurs [eIANT]e [enpisos So oo oo co Ee © ©S S Oo eo ©o ss [OAvIs 2 pues [eIAN][Oo/[eIANy]e [enpisol =a Oo moe So ©S SS © Oo = suieydpues (9}14M) snosoryIs S Oo 2S ©& sorydrouiejou a snosusi oyewe.yjn Se oo oe Se 2 © S So eo 2 © 23 RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION 5 G & z. 5 & Si ee eae ° g SLE iy ge ie. S aes = 5 2 = P 8 2 ¢ a werhinweE. cmesema s Bg oe ae By eee ob o 8 - -& Eee si eie 3. Ena sim ta Bp Bw fe en Bands E. Fy 3 cae g g < a. BF oO = 2 a a. Southern Montane Heaths Om (0) 50 6) SU 0 0) OO} 100 74S 0 0 3 0 0 Alpine Heaths 0) 0) 0) 7B al OO) OO SO Say es 0 0 0 Tableland Clay Grassy Woodlands 0: BO) iO! 4 ees 1 ] Oe-0" “0D 36) BIsh 05 o 0 0 0 New England Grassy Woodlands OE OR ORS See 2 ] OO 0) 4S a7 Oe 0 0 0 Western Slopes Grassy Woodlands Wee OO a OO OLS BON ce 4s Aen) eae) 2 0 0 Western Peneplain Woodlands PE LDP MEO HO IE RS OO) Oe Or NA Oy OM SHS} 0 0 Subalpine Woodlands OO 0 © 45 Oo 0 0) WY A wie @ 0 0 0 Temperate Montane Grasslands 0 @©@ @ WW Ss Ww I @ @ mW 39 @ 1 0 0 Semi-arid Floodplain Grasslands WW O O OD OO ve Oo OO O ODO YF YM © © 5 0 0 Coastal Swamp Forests OO OO Us Oo O OFO 7 © @® @ OF ZI 0 Coastal Floodplain Wetlands Oe Os Vil Ds Oo “ol Oy OO O 2 it © © 1 2 0 Eastern Riverine Forests A Wy O i gs Bil 4 @ @ @ 33 1 © I 1 0 0 Inland Riverine Forests Ay 6 O 6 @ 1 Oo Bg O 2 @ © 1 Oe 2 1 0 0 Inland Floodplain Woodlands My O29 D@ OY O B Oo 3 O O% BY WO OM 2B 9 0 0 Coastal Heath Swamps OO Os UO 1O-43-0 O0 OO G&G 1 O © 1 31 0 Montane Bogs and Fens oO % © s8 8 6 3S © OO DO is 8 © 4 0 0 Coastal Freshwater Lagoons Oo Oo O © O DM 3 0 oO O 5S tt GD .O 2 9 0 Inland Saline Lakes LO LOU Or OS Ose O O° S200. O--O 5 0 0 Mangrove Swamps OP Ue eS Oa A O= 0 90 SO © ONS O 0 Riverine Chenopod Shrublands no 08 0 OY 1G © 9 O O F O BG BD 2 0 0 Aeolian Chenopod Shrublands m Oo © O@ O @ O 6 © O0 3 O G @ 2 0 0 Dune Mallee Woodlands S OU Oo MY O TF O O08 O O Ft OD O © 3 0 0 Sand Plain Mallee Woodlands a OO 0 OG O © WW CO O 3 OG Oo 8 0 0 Semi-arid Sand Plain Woodlands S70 OP Or) 50. 10 A 70ae 3-0) Oh 2 0 Bl 70 3 0 0 Sydney Sand Flats Dry Sclerophyll OY Ob Oe 32 Sem, 0 0 Oh 6 .0MeoD is 0 0 Forests South Coast Heaths OF JOR 10h Rk ee TS) BOR OMe 10) 10F" 43) 50m 10F SO 30) 22 a0) Dae Gorge Dry Sclerophyll ov 8 O° Ne 6 O° See ow “G w- 4. TOD 0 0 0 orests Clarence Dry Sclerophyll Forests OO O O 4ad OM Id oO OO O wil t @O 0 0 1 New England Dry Sclerophyll Forests 0 0 0 75 9 0 0 0 0 0 14 3 O O 0 0 0 24 Proc. Linn. Soc. N.S.W., 132, 2011 Hunter-Macleay Dry Sclerophyll Forests Coastal Headland Heaths Saltmarshes Coastal Valley Grassy Woodlands Montane Lakes Southern Warm Temperate Rainforests Montane Wet Sclerophyll Forests Central Gorge Dry Sclerophyll Forests Cumberland Dry Sclerophyll Forests Southern Hinterland Dry Sclerophyll Forests Southern Wattle Dry Sclerophyll Forests Upper Riverina Dry Sclerophyll Forests Southern Tableland Grassy Woodlands Riverine Sandhill Woodlands Inland Rocky Hill Woodlands Riverine Plain Woodlands Inland Floodplain Shrublands Subtropical Semi-arid Woodlands Desert Woodlands North-west Floodplain Woodlands Gibber Chenopod Shrublands Stony Desert Mulga Shrublands Sand Plain Mulga Shrublands Brigalow Clay Plain Woodlands North-west Plain Shrublands Gibber Transition Shrublands Alpine Fjaeldmarks Proc. Linn. Soc. N.S.W., 132, 2011 sule|dpues (po) ueljoor 52 55 oyUTIBOTeO S}USUIIPos ouTIeN}so D.A. KEITH SOAISNIQUL OISTOF 55 SOIUBOIOA OISTOF WINIANT]e ure;dpooy. 78 16 11 38 Areyuaupes zyrenb ysiy Sil S]USUIPes SULSNOR] 9}119}] QUO}SOULT] Areyusutpas Zj1enb Moy 14 58 45 SOAISNIJUT 2 SOIUBOTOA OYeUI Avjo [eIAN]e [enpisor spues [eIAn][e [enpisor [OARIS 2p pues [PIANT[OO/[eIANT[e [enpisor 34 26 sure|dpues (931M) snosoiyis — 69 22 sorydiourejoul 2 snooust oyvuenin XS) RELATIONSHIPS BETWEEN GEODIVERSITY AND VEGETATION ty a 3 tn ” ea 2 _ = a = o ro) => = = ° S 8 pb & fF a 3D. eh ee =e 6 =a = 4 ie ey, eo! Be eee ple ei Pn a ee a 8 S! BiGioe eo BON boo) aie - © te ie. apy ibedsae =~ eis z es a 5 el a = 5 = mm 1s SS Re mio ec ake 2 Plies Bo) ey ee Pie SS = a 5 4 2 gc 6 =) 3 8 3 5 2 s = & sp oie ele Mare ee Ebi ie see ln > 2 Foe = 3 0 p ro) ad ¢ y wv a 6 = = 2 = 2 pa aid a g a FA g rh ee oO aL = Ee ©£ = (S Bia: A ‘ = 2 2 OS Sb Va Fe Ob sete ee te Se me Ss eas, ee Bi Oh 3B! Be Ss ee ees ae OOS teil Ae ee een ao] o Oo 9 = Oo a 2 = ° & = 5S AS — 53 7 a ‘= 3 s 5 g a 2 <= a. 77 i) = 5 = = S = 2) > > Uv 5 oe i) 5 wa 5 a < o = & w < = a an = vi : 6 BF uw = L. ray Yetman Dry Sclerophyll Forests ] O20), 36;, 40)” 32) 59) 40) SO) 80K. 26 7 OO | 0 0 North-west Alluvial Sand Woodlands 0 0 0 0 0 84 O 0 0 0 2 0 0 7 6 0 0 Inland Floodplain Swamps 25 OO a) i 89" 306 SS 40-> 40. AS ORG ] 6 0 0 Floodplain Transition Woodlands 10°70; 0) 4 23) 7 ] O° 40) AT 92 liGEary 12.750 0 Western Slopes Grasslands 3 OY @ 1 46 1 2s AO) MO) Gs Si OG) 1 0 0 Seagrass Meadows Ole 0) 40), Si 20" AOR M0) 80) Vi ORO 1 Ss’ a0 Nn ee SlopesPry Sclerophyll i. % 0 > & B90) Mo % 257 His MOC lren Alpine Herbfields 0) YO) 10), 29) a) 2 40) 80) 10) 2 221 SOG 0 0 0 Alpine Bogs and Fens OF 10) 40) 180 2 2 10 S80) 0-0) LO ro 0 0 0 Maritime Grasslands O- MOT MO} Oe) 102 5) 92> FO MOON es Oe) OP essed Oceanic Rainforests Ors O O O 2 O71 0 O O O DM © 0 0 0 0 Oceanic Cloud Forests OF Oo OO Oo oO OF & Oo O OY OD KK © oO 0 0 0 Western Vine Thickets 0 0 “© O 7 ~2 F8 40) 0 “0 “4il. SAO Xo 1 0 0 Temperate Swamp Forests 0} 0) 0) 24) Si 8) 8 0 OF OR 2S) 00 3 0 0 Riverina Grasslands tf 0 0 80 B20 Bh FOr 40) BO SO: oom 0 0 0 Wadi Woodlands 25. 10" 10) 10) 70) 389) 40) 10) FO) TS) 0) 0) Oe OO 0 26 Proc. Linn. Soc. N.S.W., 132, 2011 The Tasmanian Geoconservation Database: A Tool for Promoting the Conservation and Sustainable Management of Geodiversity MICHAEL COMFORT AND ROLAN EBERHARD Department of Primary Industries, Parks, Water and Environment; GPO Box 44, Hobart, Tasmania 7001. (Michael.Comfort@dpipwe.tas.gov.au) Comfort, M. and Eberhard, R. (2011). The Tasmanian geoconservation database: a tool for promoting the conservation and sustainable management of geodiversity. Proceedings of the Linnean Society of New South Wales 132, 27-36. The Tasmanian Geoconservation Database (TGD) is a source of information about earth science features, systems and processes of conservation significance in Tasmania. It evolved when a number of sources were compiled as a single geoconservation digital dataset as part of the National Estate component of the 1997 Commonwealth-Tasmanian Regional Forest Agreement. The latest version of the TGD (version 7) was published in 2010 and lists some 1049 sites ranging in scale from individual rock outcrops and cuttings that expose important geological sections, to landscape-scale features that illustrate the diversity of Tasmania’s geomorphic features and processes. The TGD is accessible to the public through Departmental websites. It is used as a planning tool in land management and in assessing development proposals at various scales. Under Tasmania’s three major environmental codes of practice, the TGD must be consulted and certain actions are prescribed where a TGD site is present. Manuscript received 22 November 2010, accepted for publication 16 March 2011. KEYWORDS: Database, Geoconservation, Geodiversity, Geoheritage, Tasmania INTRODUCTION Tasmania is Australia’s smallest state, and lies to the south east of mainland Australia separated by Bass Strait. It is comprised of 344 islands covering 68,401 square kilometres of which the main island occupies 62,409 square kilometres. Except for the outlying Macquarie Island, Tasmania and its islands lie between 39°14’ and 43°51’S latitude and 143°50’ and 148°29°E longitude. The isolated subantarctic Macquarie Island, located at 54°30’S 158°57’E is also part of Tasmania. Within this relatively small area lies an enormous range of geodiversity. There are geological units from every one of the 12 major periods of earth history from the Precambrian to the Holocene spanning some 4,600 million years. Geologically, it could be described as a microcosm of eastern Australia, with additional distinctive Tasmanian elements, such as extensive dolerite intrusions associated with the break-up of Gondwana. Landforms in Tasmania are also very diverse and include: rugged mountain ranges, spectacular glacial features, periglacial landforms, largely pristine river catchments, extensive limestone and dolomite karstlands, inland dunefields and a range of coastal features including a number of relic features. Soil types vary across the state and are controlled by the bedrock and a range of soil forming processes. In short, Tasmania is a very geodiverse state. Given the geodiverse nature of Tasmania, perhaps it iS no surprise that Tasmanian earth scientists have played key roles in the relatively recent field of Geoconservation (Dixon 1995, Gray 2004 and Sharples 2002). Houshold and Sharples (2008) provide a history of geoconservation in Tasmania. Sharples (2002) defined the terms geoconservation and geodiversity as follows: Geoconservation is the identification and conservation of geodiversity for its intrinsic, ecological or heritage values. Geodiversity is the natural range (diversity) of geological (bedrock), geomorphological (landform) and soil features, assemblages, systems and processes. Geodiversity includes evidence for the history of the earth (evidence of past life, ecosystems and environments) and a range of processes (biological, hydrological and atmospheric) currently acting on rocks, landforms TASMANIAN GEOCONSERVATION DATABASE Table 1. List of geoheritage inventories consulted as part of the process to compile version 1 of the Tas- manian Geoconservation Database under the Commonwealth- Tasmania Regional Forest Agreement. Year Inventory Reference 1979 Geological monuments in Tasmania Eastoe (1979) 1987 Geomorphological Reconnaissance of the Southern Forests area, Kiernan 1987 Tasmania 199] Earth Resources of the Tasmanian Wilderness World Heritage Area Dixon (1991) 1993 A Preliminary Geoheritage Inventory of the Eastern Tasmania Terrane Bradbury (1993) A reconnaissance of landforms and geological sites of geoconservation 1994 significance in the North-Eastern Forest District (Eastern Tiers and Bass Sharples (1994) Forest Districts) Continuation of Preliminary Inventory of Sites of Geoconservation ne Significance in Tasmania Central, Northern and Western Tasmania eae) 1995 Geomorphological Reconnaissance of the Southern Forests area, Kiernan (1995) Tasmania A reconnaissance of landforms and geological sites of geoconservation 1995 significance in Eastern Tasmania (parts of Derwent and Eastern Tiers Sharples (1995) Forest Districts) 1996 Inventory and management of Karst in the Florentine Valley, Tasmania Eberhard (1996) 1996 A reconnaissance inventory of sites of geoconservation significance on Dixon (1996) Tasmanian islands A reconnaissance of landforms and geological sites of geoconservation no significance in the Murchison Forest District SHEDS (122102) 1996 A reconnaissance of landforms and geological sites of geoconservation Sharples (1996b) and soils. These definitions have been adopted in Tasmania and the concepts of geoconservation and geodiversity are considered an integral part of nature conservation within Tasmanian land management authorities. With the recognition of geoconservation in Tasmania, a tool to assist in the management of Tasmania’s significant geoconservation features was required and the Tasmanian Geoconservation Database (TGD) evolved. Details of the TGD, its history, structure and uses are described below. There are a number of different approaches to managing information about geoconservation values within other Australian States, however it is beyond the scope of this paper to assess or compare these. Development of the Tasmanian Geoconservation Database The TGD was developed as part of the process leading up to the 1997 Commonwealth - Tasmania Regional Forest Agreement (RFA) under the comprehensive regional assessment (Dixon and Duhig 1996). This process enabled the compilation of a single digital database of significant geoconservation sites across Tasmanian. In generating the list of sites, a 28 significance in the Circular Head Forest District number of documents already listing geoconservation values across Tasmania as a whole or dealing with specific regions of the state were consulted. The earliest of which dated back to 1979, when the Geological Society of Australia published a report on Geological Monuments of Tasmania — a descriptive list of fifty or so geological features and landforms (Eastcote 1979). A number of subsequent geoheritage inventories produced by the Parks and Wildlife Service and the (then) Forestry Commission in the 1990s formed a significant resource in compiling the initial database. Table | lists key inventories referenced as part of this process. The RFA process led to a database with 900 geoconservation sites. The (then) Department of Primary Industries and Water took responsibility for managing the database in 1999 and established an expert panel (see Listing Process below) to advise on the listing of sites. In 2005 a summary version of the TGD was first published on the web establishing it as a standard reference for planning and land management within Tasmania (Eberhard and Hammond 2007). The latest version of the TGD (version 7) was published in 2010 and lists some 1049 sites. Further development of the TGD, and the Proc. Linn. Soc. N.S.W., 132, 2011 M. COMFORT AND R. EBERHARD Table 2. Primary level and type site classification in the Tasmanian Geoconservation Database (Dixon and Duhig 1996 and Version 7 TGD). These fields are in- tended to illustrate those elements of the site which are significant and are not used in a purely descrip- tive manner. In classifying sites, additional types are permitted if the listed ones do not provide a relevant option. Primary level Type Geology Classical (a) Historical (b) Igneous — intrusive Igneous — volcanic Metamorphic Mineralogy Palaeoenvironment Palaeontology Petrology Relationship Geomorphology Aeolian Coastal Karst Glacial Marine Estuarine Lacustrine Periglacial Fluvial Periglacial Mass movement Weathering Erosion Surface Structural landform Organic (undifferentiated) Swamp peat Blanket bog Basalt Soil Laterite Palaeosol Duricrust Alkaline pan Mineral soils undifferentiated (a) Refers to features known from the literature or some other way to earth scientists (b) Refers to features with local historical interest additional to their geological interest. Proc. Linn. Soc. N.S.W., 132, 2011 provision of advice concerning listed sites, is core business for the Geodiversity Conservation and Management Section, part of the Tasmanian Department of Primary Industries, Parks, Water and Environment (DPIPWE). Structure Currently the TGD comprises two data sets, with textual information stored in a Microsoft® Access database and spatial information (stored as polygon data) along with a subset of the text fields in an Oracle database. A program to transfer all TGD data onto a restructured Oracle database is currently underway (see Future Directions below). The database comprises a number of fields that describe various attributes of the sites. Dixon and Duhig (1996) and Sharples (2000) describe the fields more fully. A separate spatial layer is attached to each site. Many of the fields are simple identification or broad descriptive fields (e.g. ID code, GIS code, Site name, Coordinate description, Coordinates, Size, Physical form of site etc) and are generally self explanatory. Sites are primarily classified into geology, geomorphology and soil types and are further subdivided as shown in Table 2. These fields are intended to illustrate those elements of the site which are significant and are not used in a purely descriptive manner. A site may have multiple entries where it is considered significant for more than one type or sub type. Significance, level, age, sensitivity, degradation and conservation fields are common to each of the geology, geomorphology and soils types or sub types. Each listed site is assigned a significance level on a scale that includes world, Australia, Tasmania, region, or local. These are described by Sharples (2000). The sensitivity field is a number that gives a general indication of the kinds of impacts that would degrade the value of the site. The scale is roughly linear on a scale of 1 to 10 following Kiernan (1997). A site with a sensitivity of | is very sensitive to damage, while a site with a sensitivity of 10 is robust such as large regions whose geoconservation values reside essentially in their large scale form. For all sites there is an overall significance and sensitivity field that encapsulates the most significant and most sensitive aspects of the site. A limited number of sites within the database are listed as restricted and specific site information is not available to the general public for these sites. Such sites are very sensitive and vulnerable to DD, TASMANIAN GEOCONSERVATION DATABASE physical damage or complete loss through collection. Typically localised fossil or gemstone sites fall within this class. When a web-based spatial search is done on an area where such sites occur a message will appear to inform the user that a restricted site is located in the search area and to contact DPIPWE. Public access to the database through DPIPWE websites provides access to spatial information and limited site textual information. The Department also provides full copies of the database to interested parties (typically Government agencies or large private land managers and consultants) under a standard licence agreement. Listing process Any person can nominate a site for consideration for listing on the TGD or propose an amendment to an existing TGD site. Proposals to add, delete or amend sites are assessed by an independent scientific panel. The panel, known as the Tasmanian Geoconservation Database Reference Group (TGDRG) is comprised of members with demonstrable expertise in aspects of Tasmanian geodiversity. The composition and roles of the TGDRG are defined under Terms of Reference (DPIWE 2009) and state that the TGDRG will comprise at least six persons and that the disciplines of geology, geomorphology and soil science will be each represented by at least two persons. Current membership of the TGDRG includes representatives from staff at the University of Tasmania, Tasmanian Minerals Council, government departments and independent consultants. There are currently fourteen members. Members are appointed by the General Manager, Resource Management and Conservation Division (RMC) of DPIPWE. The group generally meets annually to consider nominations and amendments. DPIPWE provides a secretary to the group. Subcommittees of the TGDRG may be formed to address specific issues and advice may be sought from non-member peers acknowledged by the TGDRG. Recommendations from the TGDRG on listing and de-listing of sites are made to the General Manager, RMC who has ultimate responsibility for the TGD. Listing criteria for TGD sites are as follows and are set out in the Terms of Reference (DPIPWE 2009). The criteria are general and provide scope for considering a broad range of values, including the more traditional geological reference sites e.g. type sections, as well as landforms and assemblages of geodiversity related values. The expert panel validates site nominations and provides scientific rigour to the listing process. 30 e Consideration will only be given to listing sites that have developed as a result of natural processes. Natural features exposed artificially (e.g. road cuttings, quarries etc) will be considered. e When listing sites, consideration will be given to the degree and clarity with which sites exhibit or exemplify the important characteristics and values of their type. e Where appropriate classificatory frameworks are available, priority will be given to the inclusion of representative exemplars of the different classes of geodiversity. e _In the absence of appropriate classificatory frameworks, priority will be given to the inclusion of the widest possible range of distinctive elements within each geodiversity theme. e The assessment will take account of the integrity of natural features and processes that contribute to site significance. Degraded sites may be listed provided they maintain part or all of their relevant geoconservation values. e Sites will be assessed according to their significance within a hierarchy of levels ranging from global to local. The assessment will consider the georegional context where appropriate. e In cases where other natural values contribute to the geoconservation significance of the site, sites may be included, conditional upon appropriate professional advice. e Sites under consideration will be deleted from the TGD if not accepted as listed sites within five years of being nominated. A nominated site must be supported by an explicit statement of significance, justifying its importance with reference to other potentially comparable sites and/or unique or distinctive elements. This information is then evaluated by the TGDRG and a recommendation made regarding the suitability of the site for listing in the TGD. The listing criteria emphasise representativeness — the degree to which a site encapsulates characteristic elements of Tas- mania’s geodiversity — in order to ensure that good examples of even common features are considered. The intent here is to ensure that commonplace features do not ultimately become rare through lack of recognition that they too contribute to geodiversity. Further work is required to develop appropriate classificatory frameworks for geodiversity to implement this goal in a comprehensive way. The listing status of new sites goes through the following stages: Proc. Linn. Soc. N.S.W., 132, 2011 M. COMFORT AND R. EBERHARD e Proposed site — sites submitted to the TGDRG, prior to being formally considered by that group. These sites are not included in published versions of the database. e _ Site under consideration — site tabled at the TGDRG, where the group determines that the site potentially satisfies the criteria for listing but requires more information before accepting it for listing in full. Sites under consideration are included in published versions of the TGD. e —_ Listed sites — sites accepted for listing by the TGD Reference Group. Implications of TGD listing The database is a resource for anyone with an interest in conservation and the environment, however, the principal aim is to make information on sites of geoconservation significance available to land managers in order to assist them manage these values. The TGD is used extensively in land use planning within Tasmania. Under present Tasmanian law, the TGD has no statutory basis and geodiversity generally lacks ' statutory protection comparable to that applicable to threatened species or Aboriginal heritage for example, which cannot be interfered with without authority, irrespective of the tenure of the land. Explicit legal protection for geodiversity is restricted to Crown reserves managed under the National Parks and Reserves Management Act 2002, which establishes the conservation of ‘geological diversity’ as a statutory management objective for reserves under the Act (evidently the term ‘geological diversity’ was adopted in drafting the legislative because ‘geodiversity’ was not defined in the Macquarie Dictionary. However, the Act indicates an essentially identical meaning for geological diversity:‘the natural range of geological, geomorphological and soil features, assemblages, systems and processes’). Under s4 of the National Parks and Reserved Land Regulations 2009, it is an offence to ‘interfere with, dig up, cut up, collect or remove any sand, gravel, clay, rock or mineral or any timber, firewood, humus or other natural substance’ or to ‘remove, damage or deface any rock, stalactite, Stalagmite or other formation in a cave’. This requirement applies to about 2,350,000 ha or 35% of Tasmania’s land mass, including many sites listed in the TGD. Notwithstanding the lack of broader statutory protection, sites listed in the TGD are subject to constraints under a variety of administrative processes. Of particular importance are three key State Codes of Practice: the Forest Practices Code (Forest Practices Board 2000), Mineral Exploration Code of Practice Proc. Linn. Soc. N.S.W., 132, 2011 (Bacon 1999) and the Reserve Management Code of Practice (PWS ef al. 2003). These documents specify acceptable standards of environmental practice during forest operations, mineral exploration and mining and reserve management respectively. They require development proponents to consult the TGD and seek expert advice on protection requirements where listed sites are present. The State Environment Protection Authority has produced guidelines to assist proponents prepare development proposals and environmental management plans for developments classified as Level 2 activities under Tasmania’s Environmental Management and Pollution Control Act 1994. S4.7.2 (f) of the guidelines requires proponents to consider ‘effects on sites of geoconservation significance or natural processes (such as fluvial or coastal features), including sites of geoconservation significance listed on the Tasmanian Geoconservation Database’. Some local government planning authorities require development proponents to address the potential presence of TGD sites on land subject to planning applications. In addition to formal requirements of this kind, the TGD has become a standard reference in virtually all contexts requiring consideration of environmental effects in Tasmania, ranging from major projects of State significance to farm dams to local government planning schemes. Its success in this regard evidently reflects growing awareness that geodiversity underpins ecosystem processes generally and must be considered alongside biodiversity in conservation and sustainable land management initiatives. A limitation of the database is that the TGD lists sites of known significance, but is not based on a comprehensive State-wide inventory of geoconservation values, and the absence of identified values at a particular location may reflect gaps in the database rather than as conclusive evidence that geoconservation values are not present. Most systematic geoconservation surveys that have been conducted to date have been based on public land based around land management boundaries (see Table 1). Sites Currently there are 1049 sites listed on the TGD (version 7). The distribution of these sites is shown in Figure 1. Sites vary in size from small individual rock outcrops and fossil sites less than one hectare, to large landscape sites of several hundred thousand hectares. The three largest sites are the: Central Highlands Cainozoic Glacial Area (781,455 ha); the Tyennan region (643,412 ha) and the Western Tasmanian 31 TASMANIAN GEOCONSERVATION DATABASE Figure 1. Distribution of Tasmanian Geoconservation Database listed sites (version 7). Macquarie Is- land, 1200 km to the south east of Tasmania is not shown nor are islands containing sites north of Flinders Island in Bass Strait (Hogan, Kent group, Curtis, and Moncoeur Island group) + O° 25080 100 150 SS) Kilometres 32 Proc. Linn. Soc. N.S.W., 132, 2011 M. COMFORT AND R. EBERHARD Table 3. World significant sites listed on the Tasmanian Geoconservation Database ( version 7 ). Sulphur Creek Pillow Lava and Folds Hellyer River Insect Fossil Locality Reward Creek Mineralisation Lemonthyme Creek Glacials Poatina Fossil Crab Site Little Rapid River Early Oligocene Plant Fossil Site Lake Fidler and Sulphide Pool Meromictic Lakes Lake Morrison Collingwood River White Schist Balfour “String of Beads’ Fossil Locality Tessellated Pavement Cape Surville Dolerite Feeder Intruding Basement Dianas Basin Folds Penguin Megabreccia Florentine Road Gordon Group Stratigraphic Sections Lords Siltstone Unit/Gordon Group Stratigraphic Sections The Fossil Cliffs City of Melbourne Bay Foreshore Upper Gordon Group Stratigraphic Sections Florentine Valley Gordon Group Stratigraphic Sections Darwin Crater Adamsfield Workings Mineralogy Rodway Valley Blockfield Lower Gordon River Levee - Flood Basin System Cynthia Bay Moraines Mt Anne (North East Ridge) Glaciokarst Lake Pedder (the original) Exit Cave - D’Entrecasteaux Valley Karst Area World Heritage Area Sandy Coasts Macquarie Island Oceanic Lithosphere New River Undisturbed Fluvial and Karst systems Weld River Basin Karst and Fluvial Systems Macquarie Graben Fluvial Geomorphic Systems Central Plateau Terrain Western Tasmania Blanket Bogs Cashions Creek Limestone/Gordon Group Stratigraphic Sections Blanket Bogs (596,637 ha). Many sites overlap one another and the total area of the state covered by TGD listed sites is about 4,105,000 ha or some 60 percent of Tasmania. 49 sites are classed as very large (>1,000 ha), 395 as large (25-1,000 ha), 335 as medium sized sites (1-25 ha) and 277 as small sites (< 1 ha). The western half of the state has a greater density of TGD listed sites. This is due in part to the more complex geology to the west and also reflects a bias in previous geoconservation inventories that have largely been confined to public lands (Table 1), with Proc. Linn. Soc. N.S.W., 132, 2011 the largest state reserves (e.g. Tasmanian Wilderness World Heritage Area) located in the west of the state. The west also contains a number of the very large landscape scale individual TGD sites. There are TGD sites representing geological ages from the Precambrian to the Holocene. Quaternary sites account for some forty per cent of listed sites. Twenty percent of sites are of Tertiary age and Triassic, Devonian, Cambrian and Precambrian sites each comprise approximately five percent of the total. Levels of significance have been assigned to most sites (27 are listed as unknown) with 15 per cent of sites considered significant at a local level, 30 percent at a regional level, 35 percent at a Tasmanian level, 12 percent at an Australian level and 3 percent at the world level. World significant sites are considered to be rare in the world and/or, by the nature of scale, state of preservation or display, comparable with examples known internationally and may be illustrative of processes occurring or having effects at a continental or national scale. The 35 world significant sites listed on the TGD are shown in Table 3. Detailed notes on a few of these sites follow by way of example of the type of information stored on the TGD. The New River Undisturbed Fluvial and Karst systems YGD site situated roughly halfway along the south coast of Tasmania is considered a site of world significance (figure 2). It includes the entire New River drainage basin from Federation Peak (source) to Prion Beach (river mouth), including the Salisbury River tributary catchment basin. It is contained within the Southwest National Park and Tasmanian Wilderness World Heritage Area. It is the largest complete source-to- sea fluvial geomorphic system in Tasmania that is entirely mantled by old growth forest, is undisturbed by contemporary human activities including land clearance, roads or walking tracks, and shows no evidence for late Holocene disturbance to fluvial processes due to former Aboriginal activity (Sharples 2003). The basin also contains extensive high-relief Precambrian dolomites and Ordovician limestones (Dixon & Sharples 1986) that are mostly unexplored but are both known to contain extensive undisturbed karst landform systems. These include 33 TASMANIAN GEOCONSERVATION DATABASE Figure 2. Photographs of selected sites from the Tasmanian Geoconservation Database (TGD). (a) Un- disturbed New River fluvial and Karst systems (photo Grant Dixon), (b) Vanishing Falls (photo Rolan Eberhard), (c) TGD listed Precambrian ripple marks, Gardiner Point (Photo R Eberhard), (d) Com- plex geology in Devonian Mathinna group sediments Maria Island coastline (photo Michael Comfort), Organic rich soils of the Western Tasmanian Blanket Bog TGD site (photo Mike Pemberton), and (f) Interview River transgressive sand sheets (photo Rolan Eberhard). extensive caves below Precipitous Bluff and at Salisbury River (limestone), Tasmania’s largest stream sink (Vanishing Falls), and a poorly documented karst system at Forest Hills (dolomite). The New River fluvial system is considered outstanding as the largest undisturbed complete source — to — sea, temperate 34 maritime climate, fluvial geomorphic system in Australia, and as such is probably comparable to the best examples globally. The presence within the undisturbed catchment of extensive undisturbed karst landform systems is an additional geomorphic value of outstanding significance at a global level. Proc. Linn. Soc. N.S.W., 132, 2011 M. COMFORT AND R. EBERHARD The fluvial and karst geomorphic systems of the New River Basin constitute benchmark geomorphic systems of outstanding universal scientific and intrinsic value by virtue of their extent and undisturbed geomorphic processes, and were assessed to be of outstanding universal value (World Heritage significance) in their own right by Sharples (2003). Another world significant site on the TGD is the Western Tasmania Blanket Bogs (figure 2). This is a large area covering much of western Tasmania and isolated pockets across other parts of the state. It covers a combined area of nearly 600,000 ha. It is the most extensive organosol terrain in Australia and the Southern Hemisphere. Blanket bogs cover undulating country but can also form on slopes of 40°. Various geological types are covered, but the best development is on infertile, siliceous substrates. The blanket bogs developed in response to high precipitation, high humidity, and low evaporation, similar to other temperate maritime areas such as Ireland. The conservation values of the site relate to the total extent and size of the organosol terrain. The site also contains various significant component features including peat mounds, subfossils and palaeosols. A number of TGD sites are found on King Island at the western end of Bass Strait including a world significant site, the City of Melbourne Bay Foreshore. The site is a shore platform and includes a section of Cambrian rocks, including sediments (sandstone, siltstone, dolomite and mixtite), pyroclastics and lavas (flows and pillows). Pillows indicate seafloor volcanism and are spectacular, with individual pillows and flows visible in plan and section. The mixtite, once thought to be a tillite, is now considered to have a non-glaciogenic density flow origin. More recent studies indicate that the site consists of superb coastal exposures of the Late Neoproterozoic Grassy Group, including glacial deposits of the Marinoan ice age, ‘cap dolostone’, shale, peperites and pillow lavas (tholeiitic and picritic), and petrologically unusual felsic intrusives. The dated volcanics (579+/- 16 Ma) and intrusives (575+/-3 Ma) provide globally important constraints on the age of the beginning of the Ediacaran Period as well as the hypothesised ‘snowball Earth’ episode. Future directions As noted above, DPIPWE is currently in the process of restructuring the existing database and combining both the textual and spatial data into a single database on Oracle software, to be housed Proc. Linn. Soc. N.S.W., 132, 2011 within the Department’s Natural Values Atlas, a web- based product for publishing information on natural values. A number of the fields of the database reflect the fact that the TGD was developed over a decade ago as part of the Regional Forest Agreement process and with developments in geoconservation principles and practice since the TGD was first developed a number of changes to the database are proposed. Some of these are related to increased software capacity and functionality while other changes are more fundamental to database fields. The sensitivity and classification fields are likely to see the most changes. The new proposed sensitivity ratings will be related to specific activities or threats and for each site there will be a number of ratings depending on the proposed activity compared to the existing database that has a generalised sensitivity rating based on a roughly linear scale. This will enable more meaningful assessments across a broader range of activities, reflecting the expanded use of the TGD in assessing developments across a range of land tenure and land use settings. The second field where a major change is proposed is the classification field. Currently sites are classified according to categories applied during the RFA based on earlier work by Dixon (1991). Despite its then innovative nature, it is no longer considered adequate for present purposes and a new classificatory system has been developed and trialled. It is expected that this will greatly improve the functionality of the TGD and enable enhanced searching functionality. Thenew software willalso enable sitenominations to be entered on-line, with various innovations to ensure more consistent and complete data entry. Once sites have been assessed by the TGDRG and approved by DPIPWE a new version of the database will be available to users directly and not as is the current practice of having to wait several months for new versions to be issued. Users of the new database will also be able to see a more comprehensive range of information relating to sites and it is hoped over time to expand this information to include photos, site condition reports and other information. It is envisaged the restructured database and operating system will facilitate its use as a standard planning reference, while freeing up existing staff resources to systematically survey and review sites based on geo themes or the new classificatory system to enhance the TGD. Access to the revised database will be through the Natural Values Atlas portal at www.naturalvaluesatlas.tas.gov.au 35 TASMANIAN GEOCONSERVATION DATABASE ACKNOWLEDGEMENTS Bronwyn Tilyard produced the map. Grant Dixon and Mike Pemberton gave permission to reproduce photos. REFERENCES Bacon, C. (1999) ‘Mineral Exploration Code of Practice, Edition 4’, Mineral Resources Tasmania, Rosny Park. Bradbury, J. (1993) ‘A Preliminary Geoheritage Inventory of the Eastern Tasmania Terrane’. Report to Parks and Wildlife Service, Tasmania. Bradbury, J. (1995) ‘Continuation of Preliminary Inventory of Sites of Geoconservation Significance in Tasmania Central, Northern and Western Tasmania’. Report to Parks and Wildlife Service, Tasmania. Department of Primary Industries Parks Water and Environment (2009). http://www.dpipwe.tas. gov. au/inter.nsf/Attachments/SSKA-7XN5XT/$FILE/ TGD%20ToR.pdf viewed 12/11/2010. Dixon, G. (1991) ‘Earth Resources of the Tasmanian Wilderness World Heritage Area: A Preliminary Inventory of Geological, Geomorphological and Soil Features’. Occasional Paper No. 25, Department of Parks, Wildlife and Heritage, Tasmania. Dixon, G. (1995) ‘Geoconservation: An International Review and Strategy for Tasmania; A Report to the Australian Heritage Commission’. Occasional Paper No. 35, Parks and Wildlife Service, Tasmania. Dixon, G. (1996) ‘A reconnaissance inventory of sites of geoconservation significance on Tasmanian islands’. A report for the Parks and Wildlife Service, Tasmania and the Australian Heritage Commission. Dixon, G. and Duhig, N. (1996). “Compilation and Assessment of Some Places of Geoconservation Significance’. Report to the Tasmanian RFA Environment and Heritage Technical Committee, Regional Forest Agreement, Commonwealth of Australia and State of Tasmania. Dixon, G. and Sharples, C. (1996) Reconnaissance geological observations on Precambrian and Palaeozoic rocks of the New and Salisbury Rivers, Southern Tasmania. Papers and Proceedings of the Royal Society of Tasmania. 120: 87-94. Eastoe, C.J. (1979) ‘Geological Monuments in Tasmania. A report to the Australian Heritage Commission’ Geological Society of Australia Inc. Tasmania. Eberhard, R. (1996) ‘Inventory and management of Karst in the Florentine Valley, Tasmania’. A report to Forestry Tasmania. Eberhard, R. and Hammond, A. (2007). Rocks and hard places: the Tasmanian geoconservation Database. Forest Practices News 8(2), 4-6. Eberhard, R. (2008) Karst and the Tasmanian Geoconservation Database, Australasian Cave and Karst Management Association Journal, 70: 25-28. 36 Forest Practices Board, (2000). ‘Forest Practices Code’ Forest Practices Board, Tasmania. Gray, M. (2004). “Geodiveristy valuing and conserving abiotic nature’, (John Wiley and Sons Ltd, West Sussex). Houshold, I. and Sharples, C. (2008) Geodiveristy in the wilderness: a brief history of geoconservation in Tasmania. In ‘The History of Geoconservation’ (Eds C.V. Burek and C.D. Prosser) pp257-272. (Geological Society, London). Kiernan, K. (1987) ‘“Geomorphological Reconnaissance of the Southern Forests area, Tasmania’. In: S. Cane (ed.) ‘An Assessment of the Archaeological and Geomorphological Resources of the Southern Forests Area, Tasmania’. Report by Natural Systems Research, Anutech Pty Ltd, Canberra. Kiernan, K. (1995) ‘An Atlas of Tasmanian Karst’. Research Report No. 10, Tasmanian Forest Research Council, Inc., 2 volumes. Kiernan, K. (1997) ‘Landform classification for geoconservation’, In R Eberhard.(ed) “Pattern and Process: Towards a regional approach to a National Estate Assessment of geodiversity’ Technical Series No. 2, Australian Heritage Commission and Environment Forest Taskforce, Environment Australia, Canberra. Parks and Wildlife Service, Forestry Tasmania and Department of Primary Industries, Water and Environment. (2003) ‘Tasmanian Reserve Management Code of Practice 2003’, Department of Tourism, Parks, Heritage and the Arts, Hobart. Sharples, C. (2002). “Concepts and principles of geoconservation’ at http://www.dpipwe.tas. gov. au/inter.nsf/Attachments/SJON-57W3 Y M/$FILE/ geoconservation.pdf viewed 15/11/2010. Sharples, C. (1994) ‘A reconnaissance of landforms and geological sites of geoconservation significance in the North-Eastern Forest District (Eastern Tiers and Bass Forest Districts)’. A report to Forestry Tasmania. Sharples, C. (1995) ‘A reconnaissance of landforms and geological sites of geoconservation significance in Eastern Tasmania (parts of Derwent and Eastern Tiers Forest Districts)’. A report to Forestry Tasmania. Sharples, C. (1996a) ‘A reconnaissance of landforms and geological sites of geoconservation significance in the Murchison Forest District’. A report to Forestry Tasmania. Sharples, C. (1996b) ‘A reconnaissance of landforms and geological sites of geoconservation significance in the Circular Head Forest District’. A report to Forestry Tasmania. Sharples, C. (2000) “Users guide to the Tasmanian Geoconservation Database’ at http://www.dpipwe.tas. gov.au/inter.nsf/A ttachments/SSKA-7XN5XT/$FILE/ TGD%20ToR.pdf viewed 12/11/2010. Sharples, C. (2003) A review o the Geoconservation values of the Tasmanian Wilderness World Heritage Area. Nature Conservation Report 03/06, Nature Conservation Branch, Department of Primary Industries, Water and Environment. Proc. Linn. Soc. N.S.W., 132, 2011 Diversity within Geodiversity, Underpinning Habitats in New South Wales Volcanic Areas FREDERICK L. SUTHERLAND!2 ‘Geoscience, Australian Museum, 6 College Street, Sydney, NSW 2010, Australia; *School of Natural Sciences, University of Western Sydney, LB 1797, Penrith, NSW 2751 (Lin.Sutherland@austmus.gov.au) Sutherland, F.L. (2011). Diversity within geodiversity, underpinning habitats in New South Wales volcanic areas. Proceedings of the Linnean Society of New South Wales 132, 37-54. New South Wales National Parks, Nature Reserves, State Conservation Areas and other reserves lie in diverse geological settings. One component, Cenozoic volcanic rocks, includes eroded basaltic fields, some representing shield volcanoes with central cores of silicic rocks. The central shields provide diverse habitats in the Tweed-Main Range, Nandewar, Ebor-Dorrigo, Warrumbungle and Canobolas areas. These shields result from deep geodynamic causes and increase in age, size and degree of erosion northwards giving systematic habitat variations. The northern Tweed structure (23—25 mya) exhibits lava aprons, erosional caldera rims, basement valley floors and an isolated central intrusive peak, whereas the southern Canobolas structure (11-13 mya) retains a general shield profile. Some basaltic fields had prolonged eruptive histories, as in Barrington Tops NP (60-4 mya). There, lavas form an incised plateau rimmed by valleys and escarpments. Similar lava fields occur in other parks and reserves, e.g. Mummel Gulf NP and Ben Hall Gap NP, but fertile basalt soils mostly promoted agricultural/forestry use. A marine park at Lord Howe Island lies on a submarine plateau cut into a 7 mya basaltic volcano. The volcanic landscapes provide scenic recreational parks and platforms for habitat studies, aboriginal history, geo-education and geo-tourism. Manuscript received 15 November 2010, accepted for publication 20 April 2011. KEY WORDS: basalt fields, central volcanoes, geodiversity, habitats, Lord Howe Island, National Parks, New South Wales. INTRODUCTION New South Wales encompasses diverse geological settings related to different times within an extended geological history from Precambrian to Recent. The different units have been subjected to a range of erosional events since the break-up of eastern Gondwana (Scheibner 1999; Branagan and Packham 2000; Veevers 2001). One component that plays a prominent role within many National Parks, forestry and conservation reserves is Cenozoic volcanic rocks. This stems from their relatively widespread distribution, particularly in eastern NSW, and contrasting erosional forms and soil development given by silicic and basaltic lithologies within them (Sutherland 1995). This NSW component is part of a more extended array of such rocks along eastern Australia (Fig.1), which also includes seamounts and island chains along the Tasman and Coral Seafloors (Vasconcelos et al. 2008). This paper aims to summarise this volcanic component in NSW where it underpins a range of habitats in National Parks (NP), State Conservation Areas (SCA), Nature Reserves (NR), Marine Parks (MP), Forestry Reserves (FR) and Aquatic Reserves (AR). Among c. 570 landscape types identified in NSW, two thirds are found in these reserves (Mitchell 2003). Photographic images will illustrate a range of these landforms and habitats that exist within their precincts. It is hoped that this survey will stimulate more detailed biological studies within these linked habitats and allow further assessments of these areas for geo-heritage values, potential geo- education themes and geo-tourist activities. Brief descriptions of these volcanic features within the main NSW parks and preservation areas (Explore Australia Publishing 2010) incorporate new dating on the rocks and some unpublished data. Updated information on the national parks, reserves, conservation areas and forestry reserves can be accessed on a range of websites, e.g. www. bigvolcano.com.au/; www.environment.nsw.gov.au/; www.nationalparks.nsw.gov.au/. A progress report 38 20°S 25°S 30°S 40°S GEODIVERSITY AND HABITAT IN VOLCANIC AREAS Peak Range Springsure &. Buckland | ee NSW ? Nandewar Warrumbungle 145°E Sten” Noosa Glasshouse f, ru Flinders om = Tweed Belmore Ebor Comboyne Tasman Sea East Australian Hotspot 150°E 155°E Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND lists geoheritage values for some of the volcanic holdings (Osborne et al. 1998). GEOLOGICAL SETTING The Late Cretaceous-Cenozoic volcanism that created the range of remnant land forms now exposed in New South Wales was similar to that now seen in active volcanoes observed in other within- plate basaltic areas such as the Hawaiian Islands. Volcanic activity would have ranged from relatively calm effusions and lava fountaining, through more continuous gas blasting of larger ejected blocks and in some cases more extreme explosive activity forming towering Plinian-style eruptive columns (Parfitt and Wilson 1995, 1999). Lava flows ranged from blocky to ropy forms that could encase internal drainages of lava and extend into long lava flows (Cashman et al. 1998; Sheth 2003). As in Hawaii, some of the volcanoes developed large shields over deep magma chambers (Kauhikau et al. 2000) from which more evolved silicic rocks could rise into their . summits (Bohrson and Reid 1997; Van der Zander et al. 2010). Such volcanos are called central volcanoes in eastern Australia and the Tasman Sea; in similar fashion to their Hawaiian and other counterparts they show a progressive increase in age away from a deep fixed mantle ‘hot spot’, as the overlying plate moved across the melting zone (Duncan and McDougall 1989; Vasconcelos et al. 2008). These linear chains of central volcanoes show some gaps and bends in their paths, which are related to further deep geodynamic processes or crustal collisions (Sutherland 2003; Knessel et al. 2008). In inland NSW, several minor volcanoes formed of a potassic lava leucitite also formed a linear age chain related to Australia’s northward movement (Cohen et al. 2008). These Figure 1. LEFT, Eastern Australia, showing re- lationships of NSW volcanic fields to the overall Cenozoic volcanic distribution. Central volcano fields (black areas) are named as major centres (inland, bold italics; coastal. non-italics) and are shown in relation to a present East Australian hotspot position. Basalt fields (grey areas) are des- ignated in NSW by symbols for the main fields de- scribed in this study (CNE Central New England; NB North Barrington; W Walcha; LR Liverpool Range; BT Barrington Tops; CC Central coast; BM Blue Mountains; A Abercrombie, SH South- ern Highlands; S Snowy ; M Monaro). The dia- gram is adapted from Cohen et al. (2008). Proc. Linn. Soc. N.S.W., 132, 2011 leucitite volcanoes do not include significant parks or reserves and are not considered further in this paper. The majority of volcanoes in eastern Australia are basalt-only lava fields. These volcanoes are less clearly related to Australia’s northward plate motion and were erupted in sporadic bursts from c. 100 mya to near-recent times. The main New South Wales volcanic fields discussed in this paper show differences in age distribution between the central volcanoes and basalt lava fields (Fig. 2). General ages of basalt lava fields and central volcanoes in NSW based on K-Ar dating and the relationship of the central volcano trend to past plate motions of eastern Australian from 90 mya to the present are depicted in Fig. 3. Where more reliable dating of the rocks is available using the *°Ar/ *Ar method (Cohen 2007), it is designated as Ar-Ar dating in this account. The contrast in the compositional ranges for typical rock types found in central and basaltic lava field sequences is illustrated using two examples from northern NSW (Table 1). CENTRAL VOLCANO COMPONENT Tweed Volcano This is the largest central shield (80 x 100 km across) and straddles the NSW-Qld border region (Duggan et al. 1993). Its growth is now dated as at least 24.3 + 0.4 to 23.1 + 0.2 mya (Ar-Ar dating; Knessel et al. 2008), although the full stratigraphical range of lavas from Tweed remains to be analysed (B. E. Cohen, pers. comm. 2010). Progressive erosion of the original structure (Willmott 2003, 2010) has reduced its landscape to (1) remnant basaltic lava aprons on its northern, western and southern sides, (2) escarpments where an ‘erosional’ caldera occupies the valley floors of the Tweed River systems (Fig. 4a) and (3) a prominent isolated central peak (Mount Warning), with some surrounding ring dyke protrusions, left by the more resistant intrusive conduit of the volcano (Fig. 4b). The highest remnant lies at 1175 m asl and the flows extend to below sea level. Tomewin Rock on the NSW border is a coarse rhyolite agglomerate that seems to represent an initial violent phase of the Tweed Volcano (Willmott 2010).The basaltic apron does not extend south into the Alstonville-Ballina area where older (27-41 mya) flows are exposed (K-Ar dating; Cotter 1998). The Border Ranges NP, Wollumbin NP (including Mount Warning NP), Mebbin NP, Nightcap NP, Whian Whian SCA and Cook Island AR are areas where views of remnant rocks of the Tweed volcano are encountered. Mount Warning is also named ‘Wollumbin’ an aboriginal 39 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS (3- 73) Ebor et (69472) Walcha Comboyne Ms-s9) 16-18 Barrington | Range small basalt remnants (13- 100) © Newcastle (SYDNEY Figure 2. Distribution of central volcanoes (stippled areas) and basalt lava fields (black areas), Qld-NSW (26—-34°S), showing ages in mya (Ar-Ar ages, no brackets; K-Ar ages, brackets), general central volcano progressive age trends (arrowed lines) and western edge of the Sydney Basin (hatched line). The Belmore central volcano is outlined by a box to indicate its unusual silicic nature. The diagram is modified from Sutherland et al. (2005b). name for a fighting chief, although some applications of aboriginal place names and history in the area are controversial. One elder source maintains the peak and adjacent caldera was known as ‘Walambing Momoli’ by the Ngarakwal people, which described its silhouette as a scrub turkey and its nest (Boileau 2006). 40 Many of the parks and reserves within this volcanic apron form part of the Gondwana Rainforests of Australia World Heritage Site (UNESCO 2010). The geology, characteristic land forms, typical soils and vegetation regimes of these areas are listed for the North Coast subregions (www.environment.nsw. gov.au). The Border Ranges NP, because of its many Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND NSW Volcanic Ages | @ Basalt age ®¢ Central volcano age | A Leucitite age | X Zircon FT age Plate Movements EF. Australia ; : | aie ian = 3 i ™~ : ~~, Central volcano trend 140° 145° 150° 155° “. Zircon age trends a — c=) o cn — 4) tn a] Figure 3. Left side: Age (K-Ar)-Latitude plots for NSW basaltic fields (filled spaces), central volcanoes (filled circles), leucitite fields (filled triangles), zircon fission track eruptive reset ages (crosses) and pro- gressive age trends (arrows). The diagram is adapted from Sutherland (2003). Note that the Central volcano trend (arrow) would differ slightly in position using Ar-Ar dating (open circles trend). The cen- tral volcano age trend (arrow head) is related to a present East Australian plume line positions at 0 mya (0 Ma line), shown in the right hand side map. Right side: Plate movement map showing past plume line positions (coloured circles with tracks) reconstructed at increasing 10 mya intervals northwards from 0—90 mya. The past positions are based on an Indian-Atlantic Ocean hotspot reference frame (1); one track (purple circles) is based on a Pacific hotspot reference frame for comparison (Maria Seton, University of Sydney, plate movement program). The Coral Sea Ridge (CSR) and Cato Trough (CT) spreading ridge system (double line) and triple point positions (stars) are shown relative to a 65 mya position Proc. Linn. Soc. N.S.W., 132, 2011 4] GEODIVERSITY AND HABITAT IN VOLCANIC AREAS Table 1. Comparative compositional ranges for some NSW volcanic fields Compositional ranges are summarised from cited literature and earlier listed references Rock Type SiO, ALO, Total FeO MgO Na,O k,O Tweed-Focal Peak central volcano sequence (significant silicic component) Basanite suite 44.5-46.8 14.7-15.1 12.4-11.3 8.7-9.0 3.6-3.8 0.8-1.5 een peel 45.6-47.5 185-163 10.8-13.3 4.5-10.2 7 0.7-2.0 Wea oe, 47.9-48.1 616.9 9.0-12.2 4.7-4.9 3.9-4.1 5-16 basalt suite Res Basalt Ope ssi 14.4-15.4 8.9-12.9 2.4-8.1 3.3-4.2 015-3.6 Silicic suite 58.5-74.3 11.3-15.0 S500 one28 2.7-5.8 1.9-4.8 Liverpool Range basalt field sequence oe feel 45.6-47.7 14.9-16.1 9.9-10.2 10.3-11.2 2030 1.5-1.8 UNSW CEE ae GIS | HOF 9.6-11.1 22-94 0.9-1.5 suite distinct landscape habitats in a relatively small area, has the highest concentrations of marsupial species and among the highest concentrations of bird, reptile, amphibian and bat species in Australia. It has particular interest in representing a transition between northern tropical and southern temperate faunal regions. The Lost Wilderness FR within the area includes over 60 threatened plant species. Nightcap NP with its eroded basalt and rhyolite landscape includes significant faunas such as the little-bent winged bat, woompoo fruit dove, masked owl, Stephens banded snake and red legged pademelon, while Whian Whian SCA within the park protects quoll, koala and platypus habitats. Cook Island AR incorporates a basalt pedestal as an important breeding ground for migratory birds and protects surrounding off shore marine reef communities. Main Range-Focal Peak Volcanoes This extended volcanic complex (80 x 80 km) west of Tweed volcano is largely exposed in Qld where the youngest basalt cap lies at 1156m asl (Stevens and Willmott 1996, 1998), but its most southern basaltic and silicic parts overlap into NSW (Thompson 1974). The Focal Peak volcano is overlapped by the Tweed lavas but rhyolite plugs assigned to it extend south to Nimbin (Willmott 2010) The Ar-Ar ages for the 42 Qld sector range from 26.4 + 0.4 to 20.7 + 0.5 mya, which suggests a wider age span than for the Tweed volcano, but the NSW exposures remain undated. The Nimbin Rocks are rhyolite peaks that mark a sacred aboriginal site named after ‘Nyimbunji’, a ruler of supernatural powers (Tacon 1998). The Border Ranges NP includes basalt lavas and some rhyolite plugs, such as Mount Glenie, erupted from the Focal Peak Volcano. Toonumbar NP, Richmond Range NP and Mallanganee NP extend across the more southern eroded remnants of the Focal Peak centre. Toonumbar NP, with peaks such as Dome Mountain, contains World Heritage listed rainforests, where unlogged tree species have been compared with those in surrounding logged areas (Kariuki et al. 2006). The habitats provide protection for threatened animals such as the sooty owl, red-legged pademelon and yellow-tailed glider. Richmond Range NP, which incorporates peripheral basalt flows from Focal Peak Volcano, includes the World Heritage listed Cambridge Plateau and holds an astounding diversity of flora and fauna, with many rare and endangered species. The Koreelah NP, Mount Clunie NP, Tooloom NP and Toobin NP lie within the less investigated NSW volcanic remnants to the west. Tooloom NP follows basalt ridges, includes the World Heritage listed Tooloom Scrub, and is a critical haven Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND POINT LOOKOUT Figure 4. Erosional features developed on older central volcanoes. (a) Border Ranges escarpment, Tweed Volcano. (b) The Mount Warning intrusive complex, from south western lava apron. (c) Governor Bluff, Nandewar Volcano. (d) Sawn Rocks (with columnar jointing), Nandewar Volcano. (e) High altitude per- spective of Ebor Volcano partially eroded on the eastern side, forming escarpment (modified from Cohen 2007). (f) Silicic dykes (light coloured) of the Ebor Crescent Complex. Photos: Benjamin Cohen. Proc. Linn. Soc. N.S.W., 132, 2011 43 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS for a wide range of threatened wallaby, potoroo, bettong, owl and lyrebird species. Belmore Volcano This small central voleano (15 x 20 km), north east of the Clarence River and east of the escarpment, is predominantly silicic in nature without significant remnants of a main basaltic apron (Sutherland et al. 2005b). The lack of basalts is unlikely to represent an erosional event, as only one basalt dyke (post- silicic) was found in the eroded interior. Three silicic rocks are dated at 20.8 + 0.8 to 21.2 + 0.3 mya (Ar-Ar dating; Knessel et al. 2008). The highest summit lies at 516 m asl and the lowest remnants lie at c. 200 m asl. Mount Neville NR (www.environment.nsw. gov. au) overlaps an outlying flow ridge from a peripheral vent of the volcano and protects plants such as spike- rush and cabbage tree species at the limits of their geographic ranges. Nandewar Volcano This central voleano (30 x 50 km) is exposed in the Mount Kapatur NP (Dawson et al. 2004) and has received detailed petrologic investigations and comment (Duggan et al. 1993; Nekvasil et al. 2004). Limited Ar-Ar dating gives ages of c. 18.5—19.0 mya for the main complex (Cohen et al. 2008). In contrast to the Tweed volcano, much of the central silicic eruptive super structure remains, reaching 1510 m asl (Fig. 3c,d), although the shield is partly dissected by radial drainage which descends through the basalt lavas at plains level (Bob and Nancy’s Geotourism site, 2010). The volcanic landforms include outstanding examples of tiered lava terraces, such as Lindsay Rocks, a spectacular set of circular dykes at Mount Yalludundida and a superb example of cooling joints in silicic lava at Sawn Rocks (34d). Mount Kapatur NP is foremost among Australian conservation areas for the range of vegetation climes that ascend its volcanic slopes over such a short distance. The varied habitats protect a diverse range of plant communities and threatened species of bats, birds, wallabies, quolls and a unique pink slug. The preserved biological communities exhibit both western slopes and tableland affinities within the area and overlaps between both northern and southern species distributions. Ebor-Dorrigo Volcano This volcano (40 x 60 km) straddles the present escarpment, producing striking topographic differences across its eroded structure from its high point at 1562 m (Fig. 3e). The volcano formed between 19-20 mya (Ar-Ar dating; Ashley et al. 1995; 44 Knessel et al. 2008). Only the western and northern basaltic aprons show substantial preservation, leaving a decapitated intrusive complex in its eroded centre (Fig. 3f) and a few residual lava caps to the south. A basalt cap at Andersons Sugarloaf (c. 850 m asl), 35 km south of the intrusive core is probably a remnant lava flow from the volcano as it is geochemically similar to analysed Ebor basalts (F.L.Sutherland and .T. Graham, unpublished analyses). This prominent peak marked a sacred aboriginal initiation site (Kempsey Heritage Inventory; www.kempsey.nsw. gov.au). The New England NP encompasses volcanic relicts left by escarpment retreat under erosion by the developing Bellingen, Nambucca and Macleay river systems. Guy Fawkes NP includes the west flowing plateau drainage, now entrenched in the basalt apron at Ebor Falls. Dorrigo NP includes some of the basalt apron on its northeastern side and with New England NP they form part of the World-Heritage listed Gondwana Forests of Australia designed to protect stands of Antarctic Beech. The rich basalt soils and wet climate support an exceptional biodiversity. Snow gum woodland, forest and heath on the high plateau pass into towering eucalypt forests and lush rainforest on the slopes. Warrumbungle Volcano A central intrusive complex features in this volcano (50 x 80 km), where erosion has exposed spectacular examples of flows, dykes, plugs and domes (Fig. 5 a, b). These showcase a wide spectrum of alkaline rocks (Duggan and Knutson 1993; Duggan et al. 1993; Ghorbani, 1999, 2003). Several Ar-Ar dates indicate that the structure developed from 18 to 15 mya (Cohen et al. 2008). Studies of minerals in the rocks reveal a complex evolution of subsilicic to silicic lavas tapped from deeper mantle and higher crustal chambers (Duggan 1990; Ghorbani and Middlemost 2000). The capping flows reach a high point at 1206 m asl and the eroded intrusive complex is readily accessible in Warrumbungle NP (Whitehead 2009). Peripheral basalt flows mark former radial drainages and descend onto the surrounding plains. ‘“Warrumbungle’ is the name given to these peaks by the Gilmaroi aboriginal people, meaning ‘crooked mountains’. Warrumbungle NP, although containing similar volcanic rocks and features to Kaputar NP, shows subtle differences in its landforms and biodiversity to its northern counterpart. The complex incisions within the volcanic edifice produce many diverse microclimates and habitats. Marsupial species abound and include the threatened brush-tailed wallaby, Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND Lord Howe Bali's Pyranud | ; a (559m) | Po isee levety Ma Gower (87 Fri} - Island _— = £000 m mann’ Old Pew 7 31,20°S Figure 5. Erosional features developed on younger central volcanoes. (a) View across eroded intrusive core, Warrumbungle Volcano. (b) Silicic plugs and dykes, Warrumbungle Volcano. (c) Basalt Plateau, Comboyne Volcano. (d) Big Nellie silicic plug, Comboyne Volcano. (e) Silicic summit from edge of basalt apron, Canobolas Volcano. Photos: Benjamin Cohen. (f) Submerged basalt pedestal, reef growth and island peaks, Lord Howe Volcano (adapted from Hill et al. 2001). bird species flourish, including a remarkable range of parrots and many lizard and snake species dwell among the rocky exposures. Comboyne Volcano This volcano (25 x 35 km) is preserved as a dissected plateau beside the main escarpment, centred near the town of Comboyne (Pain and Ollier 1986). Lower basalt flows are capped by silicic flows Proc. Linn. Soc. N.S.W., 132, 2011 between c. 400-700 m asl (Fig. 5 c) and scattered silicic intrusives up to 865 m asl (Fig. 5 d) mostly rise through basement exposures (Knutson 1989). To the southeast, lower basalts and a trachyte outcrop at Mount Juhle and further south silicic intrusives continue as far as Wingham down to c. 10 m asl. Silicic units give Ar-Ar dates from 16.5—18.1 mya (Knessel et al. 2008; F. L. Sutherland, I. T. Graham and H. Zwingmann, unpublished data). Silicic peaks 45 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS at Mount Coxcomb, Mount Goonuk and at Big Nellie, Flat Nellie and Little Nellie feature at Mount Coxcomb NR, Mount Goonuk NR and Killabakh NR and Corrabakh NP (Evans 2001; Westerman 2004), while basalt flows on Camboyne plateau feature in Boorganna NR. The name Comboyne is derived from an aboriginal word for kangaroo. The plateau-escarpment connection in_ this volcanic area provides complex habitats. Corrabakh NP is an important area for many rare, threatened and endangered plant species, with some lying at their southern limits, while the plugs at Big Nellie support eucalypt species at unusually low altitudes. Endangered animals include the bush curlew and the giant barred frog. Canobolas Volcano This small shield (30 x 50 km) largely retains a compact cone-like profile (Fig. 5 e), rising from c.900 m to a summit at 1395 m asl. Its geomorphic features, ranges in basaltic and silicic rocks and the soils are described by Pogson and Watkins (1998) and Chan (2003). Some Ar-Ar dating suggests construction from 13.3 to 11.5 mya (Cohen et al. 2008). The main edifice lies within Mount Canobolas SCA. The mountain name comes from the Wirudyri aboriginal words “Gaahna Bulla’ meaning two shoulders, which describes the two main peaks of ‘Old Man Canobolas’ and ‘Young Man Canobolas’ in the eroded volcano. Mount Canobolas SCA, located on an isolated rocky ‘island’ rising from surrounding plains, forms an important moist micro-climate habitat for plant and animal communities. Its outcrops host a variety of mosses and lichens, including endangered lichen communities. The mountain supports snow gum sub- alpine woodlands, including the threatened endemic Eucalyptus canobolas. Lord Howe Volcano This oceanic volcanic island, with its satellite Balls Pyramid to the south, falls under NSW jurisdiction and UNESCO World Heritage listing (Hutton 2008; UNESCO 2010). It is described here with the central volcanoes as part of an age- progressive oceanic volcanic chain (Mortimer et al. 2010), which has a similar, but not contemporaneous, origin to those along the eastern Australian seaboard (Duncan and McDougall 1989). It is largely basaltic, without observed silicic components, but most of the structure (Fig. Sf) forms a large, hidden submarine pedestal (30 x 80 km). Only part of its former caldera lava-filling now stands above sea level and reaches up to 875 asl (Thompson et al. 1987; Hill et al. 2001). The K-Ar dating suggests a 6.5—7 mya construction age. 46 The Lord Howe Island State MP, gazetted in 1999, and the Lord Howe Island (Commonwealth Waters) NP, proclaimed in 2000, cover several specific areas on the bevelled submarine platform on the volcano, which are presently under revised management arrangements (www.mpa.nsw.gov.au; www.environment.gov.au). Studies of the offshore marine platform recently revealed that a much larger fringing coral reef existed around the Island prior to growth of the present reef since 7 kya (Woodroffe et al. 2010). BASALT LAVA FIELD COMPONENT Significant basalt-only lava fields extend throughout eastern NSW and their soil types and vegetation show differences related to their regional climates (Jenkins and Morand 2002). The basalts range from alkaline (nephelinites, basanites, alkali basalts, hawaiites, mugearites) into subalkaline (transitional basalts, olivine tholettes, quartz tholeiites) types (O’Reilly and Zhang 1995; Vickery et al. 2007). The central New England and Walcha fields occupy significant areas of the New England Tablelands. Voluminous basalts form the Liverpool Range between the Tablelands and the Hunter Valley, while the Barrington province extends through the Mount Royal Range into the Barrington Tops plateau. Lavas in the central New England field (70 x 240 km) reach elevations over 1370 m asl, show a wide age range (14-40 mya) and include alkaline and subalkaline basalts (Vickery et al. 2007). The exposures are largely devoted to pastoral and gem mining pursuits (sapphires and zircon) and only support limited nature reserves (Glen Innes-Guyra basalts, www.environment.gov.au). A feature of some basalt fields is their growth by repeated eruptions over an extended period, e.g. for over 55 my in the Barrington province. This aspect of ‘hydra-head’ growth through progressive cut back of earlier volcanoes and subsequent replacement during the eruptive history is illustrated for the North Barrington-Barrington Tops fields in Fig. 6. Walcha field Basalts extend over 60 x 60 km and descend from 1200 m to 950 m asl. Alkaline to subalkaline types range in age from 35-73 mya and include gem-bearing types (Sutherland and Barron 2003; Sutherland et al. 2005; Gibson 2007; F.L. Sutherland, I.T. Graham and H. Zwingmann, unpublished data). Deep incision of flows feature in Mummel Gulf NP and basalts extend through Riamukka SF to the west and Einfield SF to the east. Further basalt areas lie Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND Latitude-Age Barrington Basalt Province (N-S section) aL.5 31.6 aL7 31.8 ss aN AN n\ 31.9 32.0 32.1 32.2 32.5 A 4 A ‘AN 4 Ah hh 30 40 60 70 80 Mya - Figure 6. Latitude (°S)—Age (mya) diagram for eruptive centres across a N-S section , Barrington ba- salt province, including Mount Royal Range and Barrington Tops Plateau, with lava-dominant volcanic centres (K-Ar, solid triangles) and zircon-ages (FT, open triangles). Age data come from Sutherland and Fanning (2001), Roberts et al. (2004), Sutherland et al. (2005a) and Gibson (2007). in Nowendoc NP and Ngulin NR. Mummel Gulf NP protects extensive old growth forests, which support a large range of bird species, forest bats and small mammals, such as the brown antechinus. Liverpool Range field This large basalt field (SO x 120 km) lies between 650-1400 m asl and its dating and petrology is summarised in Schén (1989). Older more alkaline lavas occur to the east (32—35mya) and younger alkaline to transitional subalkaline lavas (38-40mya) form the western sequence. Wallabadah Rock is an unusual isolated rhyolite plug chemically similar to, but older than, the rhyolites in other central volcanoes, as it gave a 46 mya age (Gibson 2007). Coolah Tops NP extends across the main erosional crest of the western basalts; it forms an isolated basalt plateau that preserves tall open forest communities that differ from the forests on other basalt reserves in the district (Binns 1996). Towarri NP on the southern basalt slopes overlaps three biogeographical regions, the NSW northern Tablelands, Briglow black soil country and Sydney Basin sandstone exposures; these habitats along with Hunter Valley acting as a conduit for migrating species hold considerable Proc. Linn. Soc. N.S.W., 132, 2011 biodiversity (Hill et al. 2001). Ben Hall Gap NP lies on the eastern basalt plateau across a drainage divide and has outstanding tall old growth eucalypt forests developed on the thick, nutrient-rich basalt soil (Mitchell 1990); it marks the northern limit of the southern cold temperature rain forests and overlaps the eastern and western distributions of many bird species. Barrington field These basalts extend east of the Hunter Valley (Chambers 1995; Sutherland and Graham 2003). The volcanoes show a wide age range (21-61 mya; Gibson 2007), with evidence of limited late activity extending to < 4 mya (Sutherland et al. 2005a) and the basalts are largely alkaline with minor subalkaline types (Sutherland and Fanning 2001). Many Barrington Tops eruptive events carried up gemstones (ruby, sapphire and zircon), which were concentrated in alluvial deposits (Roberts et al. 2004). The main basalt regions lie within the Mount Royal NP, Barrington Tops NP and Barrington Tops SCA in areas which are monitored using vegetation surveys (Zoete 2000). At Barrington Tops, the plateau basalts (Fig. 7 a) reach up to 1576 m asl, and radial drainage has developed 47 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS Figure 7. Barrington Tops basalt field, showing erosional features. (a) Plateau surface with drainage headwaters between older 59 mya basalts (left) and younger 50—55 mya basalts (right), Hunter Springs. (b) Dissected plateau scarp, northeastern Barrington Tops Plateau, looking from Moppys Lookout. Pho- tos: F.L. Sutherland. peripheral escarpments and deep valleys (Fig. 7 b) that cut through the basalts into basement rocks between 600-900 m asl. Barrington Tops NP and Mount Royal NP include segments of the World Heritage-listed Gondwana Rainforest of Australia for their subtropical rainforests that occupy valleys in the basalt plateaus (UNESCO 2010). The diverse habitats provide refuge for threatened animal species such as the Hastings River mouse. Mid-NSW basalt fields Scattered remnants extend through the Sydney- central coast area and westwards into the Oberon and Bathurst areas. Basalt dykes intrude coastal sections between Newcastle and Wollongong and some are Cenozoic, such as those at Era Beach in Royal NP south of Sydney, which gave a 51 mya K-Ar age (Och et al. 2009). Younger alkaline basalts (14-21 mya), such as at Mount Banks, fall within Blue Mountains NP (Alder and Pickett, 1997; Van der Beek et al. 2001) and Mount Yengo NP (Mount Warrenga, Gibson 2007). Older basalts (34-57 mya) extend through Mount Yengo NP, Wollemi NP and Nullo Mountain SF. They rise to 1154 m asl at Tayan Pic, a designated significant geological site, and include Mounts Coricudgy, Pomany, Corriday, Mondilla, Coorangoola and Kerry and Nullo Mountains 48 (Gibson 2007). Mount Yengo formed a significant mythological feature for surrounding aboriginal tribes as a Creator God, Biamie. These basalt peaks and soils influence local habitats within the wide range of eucalypt species developed across the sandstone platforms of the Greater Blue Mountains World Heritage area (UNESCO 2010). Basalt and dolerite remnants in Abercrombie River NP represent former lavas that extended into headwaters of west-flowing paleodrainage systems, while other basalts entered former Lachlan and Macquarie River courses downstream as far west as the Dubbo-Orange area (Bishop and Brown 1992; Tomkins and Hesse 2004). Southern NSW fields The Southern Highlands (100 x 110 km) and Grabben Gullen fields (30 x 40km) southwest of Sydney contain scattered basalt patches with diverse ages (20-60mya; Gibson 2007) and are mostly alkali basalts with confined flow extents (O’Reilly and Zhang 1995). Minor remnants lie within Tarlo NP and rainforest on basalt is preserved in Robertson NR. Further south, alkaline and subalkaline basalts form flows (40—S5Omya) within the Shoalhaven catchment area and were used to demonstrate the relative antiquity of the plateau surface (Nott et al. 1996). Some flows are found in Morton NP and Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND Budawang NP. Along south coastal NSW, similar basalt types show younger ages (27-34 mya; Brown 2000), but have induced differing interpretations of their geomorphic relationships with highland development. The northern part of Eurobodalla NP is dominated by basalts, but the largest basalt body forms Mount Durass within the bounds of Greater Murramarang NP (Wright 1996). To the southwest, basalts in the Snowy field (60 x 80 km) are mostly alkaline types that remain as valley filling ridges and plateaus of Miocene flows (18-24 mya) that range in their elevations from 450 m to summit sources up to 1784 m asl (Sutherland et al. 2002; Sharp 2004). These basalts have engendered considerable discussion on their relationships to the age and uplift history of the surrounding uplands (Young and McDougall 2004). Many of the higher. basalts are included in Kosciuszko NP and the Tabletop wilderness area while lower plateau basalts lie within Bago SF. The largest southern field (45 x 110 km) is the Monaro field, where alkaline and some subalkaline sequences (34-58 mya; Gibson 2007) preserve important evidence of early vegetation and - climatic records (Taylor et al. 1990; Brown 1994; Roach et al. 1994; Taylor and Roach 2003; Sharp 2004). Although some basalts are located in South East Forests NP and some central-northern reserves (Conartha NR, Myall NR), the bulk of the basaltic soils support grasslands that include a number of preservation areas for endemic species (Garden et al. 2001; Benson 2003). DISCUSSION New South Wales is well-endowed with national parks and ancillary reservations, containing over 380 listed sites covering 7% of the State’s area (NSW Government websites). Among some sixty national parks, most lie on the eastern side of the state (Explore Australia 2010), where some 30 of them cited in this survey contain exposures of Cenozoic volcanic rocks. This highlights the important role that this unit occupies within the geodiversity on offer in NSW reservations. The most diverse range of rocks and landforms appear in the central volcano shields, where contrasts between basaltic and silicic lithologies lead to more pronounced differences in erosional forms and soil developments. This translates into greater variations in vegetation make up and habitats for fauna. Rainforests tend to develop on nutrient-rich basalt areas while eucalypt sclerophyll stands tend to colonise nutrient-deficient soils on silicic rocks. Nightcap NP, which has the highest annual rainfall Proc. Linn. Soc. N.S.W., 132, 2011 in NSW, supports subtropical rainforest on its basalt soils and warm-temperate rainforest on its rhyolite- based soils. The lithological nature and topography in the volcanic areas also dictates land use. Areas with rich basalt soils on flatter, accessible terrains with favourable hydrological characteristics encourage agricultural use (Brodie and Green, 2002), whereas rugged scarps are often too steep for cultivation. Juxtaposed lithological contrasts in the central volcanoes provide scenic appeal for visitors and with their biodiversity factors has led to their prevalent inclusion within parks and reservations. Systematic geodiversity The NSW central volcanoes show a general change in their ages (from 25 to 12 mya) and size (from 80 x 100 to 30 x 50 km across) with latitude southwards (28.2 to 33.4°S). This change provides a systematic base to study their geological and habitat variations, related to climate and variable length of erosion and weathering time. This general rule, however, excludes Belmore Volcano. Here, the absence of basalts led to a reduced shield area and a different erosional history and land use. The older northern volcanoes (Main Range-Tweed) provide examples of more extreme erosional relief and habitat ranges than the younger southern Canobolas Volcano, where its remaining profile lacks marked internal topographic disruptions. Habitats are largely limited by basaltic/silicic soil distribution, altitude and hydrological changes from the surrounding plains to the mountain summit. Two separate central volcano chains formed during their progressive development southwards, giving eastern (QId-NSW _ border volcanoes, Belmore, Ebor, Comboyne) and western (Nandewar, Warrumbungle, Canobolas) lines. This brought another systematic erosional factor into play, the intersections of some volcanoes by escarpment retreat towards the east Australian divide during drainage development (Ollier and Pain 2000). The Tweed and Belmore volcanoes grew onto the coastal margin so that the escarpment retreat had intersected their positions by 24-20 mya respectively, but not those of the Main Range and Ebor volcano, which are only now half removed by the escarpment inroads. The Comboyne volcano remains almost connected to the escarpment and the western centres lack the extreme division of habitats caused by escarpment intersections. Among the basalt fields, systematic differences in habitats can appear where adjoiing fields show significant age differences. In southeast NSW, the older Monaro field with more deeply developed soil profiles supports natural grasslands and forests that 49 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS contrast with less deeply weathered plateau and flow caps that remain in alpine and foothill settings in the Snowy field. Geodiversity platforms Variations in volcanic rock types, their ages, landforms and weathering characteristics all feature within this one Cenozoic unit, within the overall geological diversity in NSW. This range in lithology, soil types and dissected features, at different altitudes and geographic locations, both inland and coastal has developed a multitude of diverse habitats. This linkage provides an important platform for promoting the role of geodiversity in the environment. It provides opportunities for multidisciplinary scientific studies, geo-heritage assessments, geo-education and geo-tourism. Examples of multidisciplinary studies that use NSW Cenozoic volcanic components include landform ecological analysis (Mitchell 2003), hydro- geomorphic comparisons (Gibson 2008), aboriginal stone tool analysis (Bowdler 2005; Corkhill 2005), and archaeological appraisals (McIntyre-Tamwoy 2008). Geo-heritage listings range from individual rocks (Nimbin rhyolite, Osborne et al. 1998) to clusters of sites (basalt ridges of the Liverpool and Mount Royal Ranges, Schén 1984) and also large- scale features (Warrumbungle Volcano; Australian Heritage Commission 2010). Geo-education utilises the NS W volcanic features in varied ways, including inclusion in explanatory books and guides for recreational visitors (Blanch and Kean 1995; Alder and Pickett 1997; Gold and Prineas 1997; Ferret 2005; Whitehead 2008), more specialised visitors (Duggan and Knutson 1993; Sutherland and Graham 2003) or even extending to fanciful stories for children’s (Hutchison 2010). Educational slide sets that feature Australian volcanoes include NSW examples (Lewis et al. 1998), while documentary video films feature NSW volcanic backgrounds, particularly the Tweed Volcano, in integrations of landforms and ecology (Sutherland 2008; Warth 2009). An interactive website for school students on Australian volcanoes is maintained by Uni Serve- Science (2001). Geo-tourism is catered for by a range of web sites which list NSW volcanic attractions (www.bigvolcano.com.au) and includes special self- operating tours (Bob and Nancy’s Geo-tours 2010). Although many avenues exist to explore the Cenozoic volcanic features in NSW, there is further scope for promoting their importance both for assisting in their preservation and for exploiting their explanatory role in illustrating geological processes. The systematic differences, within and between the large central volcanoes and basalt fields, provide 50 considerable capacity for developing overarching themes that link their individual features into a grander picture. For example, although each central volcano has its own geological history as a group they can be linked to Australia’s plate motion movement away from Antarctica and the concepts of global ‘hot spot’ volcanic traces elsewhere. Likewise, individual basalt fields can be related to their place within the long evolution of volcanic activity along the Tasman margin, to concepts of landscape inversion or to how they preserve records of former biodiversity during Australia’s natural history evolution. Many of the volcanic fields can be linked into preservation of ‘Gondwana’ rainforest reserves (United Nations Environment Program-Wo 2008). Unfinished story The NSW Cenozoic volcanic record developed through to its present landscapes over a 100 my period and volcanism still remains dormant in far northern Queensland, western Victoria and SE South Australia (Johnson 2004). The NSW landforms discussed here only acquired minor, mostly explosive additions in the last 10 my, while erosion further dissected the lavas. Nevertheless, the rocks remain as valuable assets to further decipher their genesis while still generating scientific debate as to the exact causes. Increasing use of more precise dating, high-quality geochemical and isotopic analysis and well-controlled geodynamic modelling continues to furnish new insights into the origin of the volcanism (Vasconcelos et al. 2008; Di Caprio et al. 2009). The onset of extensive basaltic lava field activity, as found in eastern Australia, is correlated by some workers with global mantle warming effects without other extraneous causes (Coltice et al. 2007), whereas other workers look to additional factors such as buoyant rise of hot mantle wedges when subduction ceased as important mechanisms (Rey and Muller 2010). For progressive central volcano activity on moving plates, fixed deep mantle thermal plumes are commonly invoked, but such activity can also be explained by other means (Reitsma and Allen 2003; Finn et al. 2005). For the eastern Australian central volcanoes different interpretations of their tracks include plumes deflected by mantle processes (Sutherland 2003), plumes deflected by thick subcontinental roots (Manglik and Christensen 2006) or plumes that directly record changes in Australian plate motion (Knessel et al. 2008), but resolution of these views needs further scientific testing. Thus, the intrinsic geodiversity revealed among the NSW Cenozoic volcanic areas continues to be updated and refreshed for presentation to scientific, recreational Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND and geo-tourist audiences. This volcanic heritage can be continually worked into new concepts, such as the new approaches to geodiversity and ecosystem services (Gordon and Brown 2010). CONCLUSIONS Cenozoic volcanic remnants form significant contributions to National Parks and reserves in eastern NSW. The large central volcano sites decrease in age and size from the northern border to central NSW, causing corresponding variations in landforms and habitats. The basalt lava fields show a greater age range, lack the silicic attributes within central volcanoes, and develop wider latitudinal and altitudinal habitats. The geodiversity just within this volcanic unit provides exceptional opportunities to study detailed geological, biological and human interactions. ACKNOWLEDGEMENTS Benjamin Cohen, Earth Sciences, University of Queensland, St Lucia, Brisbane, provided the photographs of central volcanoes, which greatly helped in illustrating the paper, and encouraged the development of its themes and read the script. Val Attenbrow, Australian Museum, advised on Aboriginal legends, while Francesca Kelly helped compile the script. The Australian Museum and School of Natural Sciences, University of Western Sydney provided facilities. Constructive reviews of the paper were made by Dr Larry Barron, Sydney, Dr lan Graham, University of New South Wales and an anonymous referee. The paper is dedicated to the spirit of the former Geodiversity Research Centre, Australian Museum, which generated a range of geological studies before its closure in 2004. These included studies on the volcanic rocks cited within the present study. REFERENCES Alder, J.D. and Pickett, J.W. (1997). “Layers of time: the Blue Mountains and their geology’. (NSW Department of Mineral Resources: Sydney). Ashley, P.M., Duncan, R.A. and Feebrey, C.A. (1995). Ebor Volcano and Crescent Complex, northeastern New South Wales: age and geological development. Australian Journal of Earth Sciences 42, 471480. Australian Heritage Commission (2010). “AHC final assessment report Warrumbungle National Park (PDF)’. (Australian Heritage Data Base: Canberra). Benson, J.S. (2003). The native grasslands of the Monaro region, southern tablelands of NSW. Cunninghamia 3, 609-650. Proc. Linn. Soc. N.S.W., 132, 2011 Binns, D.L. (1996). ‘Floristics and Vegetation Patterns of Coolah Tops National Park. (NSW NPWS: Sydney). Bishop, P. and Brown, R. (1992). Denudational isostatic rebound of intraplate highlands: the Lachlan River Valley, Australia. Earth Surface Processes and Landforms 17, 345-360. Blanch, R. and Kean, V. (1995). “Bushwalking in the Mount Warning Region 2" Edition’. (Kingsclear Books; Sydney). Bob and Nancy’s Geotourism site (2010). “Geological drive across the Nandewar Volcano’. (http:// ozgeotours.110mb.com). Bohrson, W.A. and Reid, M.R. (1997). Genesis of silicic peralkaline volcanic rocks in an ocean island setting by crustal melting and open system processes: Socorro Island, Mexico. Journal of Petrology 38, 1137-1166. Boileau, J. (2006). ‘Caldera to the sea: a history of the Tweed Valley’. (Tweed Shire Council: Murrwillumbah, NSW). Bowdler, S. (2005). Movement, exchange and the ritual life in southeastern Australia. In ‘Many Exchanges: Archaeology, History, Community and the work of Isabell McBride (Eds I. MacFarlane, M-J. Mountain and R. Paton) pp. 131-146. (Aboriginal History Inc: Canberra). Branagan, D.F. and Packham, G.H. (2000). ‘Field Geology of New South Wales’. (Mineral Resources New South Wales: Sydney). Brodie, R.S. and Green, K. (2002). “A Hydrological Assessment of the Fractured Basalt Aquifers on the Alstonville Plateau, NSW’ (Bureau of Rural Sciences: Canberra). Brown, M.C. (1994). An interpretation of Tertiary landform evolution in the Monaro Volcanic Province. In ‘The Tertiary geology and geomorphology of the Monaro: the perspective in 1994’. (Ed. K.G. McQueen) pp. 30-35. University of Canberra Occasional Publication 2. Brown, M.C. (2000). Cenozoic tectonics and landform evolution of the coast and adjacent highlands of southeast New South Wales. Australian Journal of Earth Sciences 51, 273-290. Cashman, K.V., Pinkerton, H. and Stephenson, P.J. (1998). Long lava flows. Journal of Geophysical Research 103 (No B11), 27281-27289. Chambers, T.N. (1995). The Tertiary history of the Mount Royal Range. BA (Hons) Thesis, Macquarie University, Sydney. Chan, R.D. (2003). Bathurst and Forbes 1:250 000 Map Sheets, New South Wales [PDF], 5 pp. (crcleme-.org. au). Cohen, B.E. (2007). High resolution *°Ar/?Ar Geochronology of Intraplate Volcanism in Eastern Australia. PhD Thesis, University of Queensland, Brisbane. Cohen, B.E., Knessel, K.M., Vasconcelos, P.M., Thiede, D.S. and Hergt, J.M. (2008). *°Ar/*?Ar constraints on the timing and origin of Miocene leucitite volcanism in south-eastern Australia. Australian Journal of Earth Sciences 55, 407-418. SII GEODIVERSITY AND HABITAT IN VOLCANIC AREAS Coltice, N., Phillips, B-R., Betrand, H., Ricard, Y. and Rey, P. (2007). Global warming of the mantle at the origin of flood basalts over supercontinents. Geology 35, 391-394. Corkhill, T. (2005). Sourcing stone from the Sydney region: A hatchet job. Australian Archaeology 60, 41-50. Cotter, S. (1998). A geochemical, paleomagnetic and geomorphological investigation of the Tertiary volcanic sequence of north eastern New South Wales. D. App. Sci (Hons) Thesis, Southern Cross University, Lismore. Dawson, M.W., Vickery, N.M., Barnes, R.G., Tardos, V.N. and Wiles, L.A. (2004). ‘Geology integration and upgrade: NSW Western Regional Assessments: Nandewar’. (Resource and Conservation Division, Dept. of Infrastructure, Planning and Natural Resources: Sydney NSW). Di Caprio, L., Gurnis, M. and Muller, R.D. (2009). Long- wave tilting of the Australian continent since the Late Cretaceous. Earth and Planetary Science Letters 278, 175-185. Duggan, M.B. (1990). Wilkinsonite, Na,Fe**,, Fe**,, Si, ,O,,, a new member of the aenigmatite group from the Warrumbungle Volcano, New South Wales. Australia. American Mineralogist 75, 694-701. Duggan, M.B. and Knutson, J. (1993). ‘The Warrumbungle Volcano: a geological guide to the Warrumbungle National Park’. (AGSO: Canberra). Duggan, M.B., Knutson, J. and Ewart, A. (1993). ‘IAVCEI Canberra 1993 Excursion Guide: Warrumbungles, Nandewar and Tweed volcanic complexes’. AGSO Record 1993/70. Duncan, R.A. and McDougall, I. (1989). Volcanic time- space relationships. In ‘Intraplate volcanism in Eastern Australia and New Zealand’ (Compl. R.W. Johnson) pp. 43-45. (Cambridge University Press: Cambridge). Evans, T. (2001).The Lansdowne Volcanics New Reserves on the Comboyne & Lansdowne Escarpments. National Parks Journal 45(1), 6-8. Explore Australia Publishing Pty Ltd (2010). ‘Explore New South Wales National Parks’. (Explore Australia Publishing: Prahran, Vic). Ferret R.R. (2005). ‘Australia’s Volcanoes’. (Reed New Holland Publishers, Australia). Finn, C.A., Muller, R.D. and Panter, K.S. (2005). Definition of a Cenozoic alkaline magmatic province in the Southwest Pacific mantle domain and without rifting or plume origin. Geochemistry, Geophysics, Geosystems 6, DOI: 10.1029/ 2004GC000723, 26 pp. Garden, D., Dowling, P.M., Eddy, D.A. and Nichol, H.I. (2001). The influence of climate, soil and management on the composition of native grass pastures on the central, southern and Monaro tablelands of New South Wales. Australian Journal of Agricultural Research 52, 925-936. Ghorbani, M.R. (1999). Petrology and Geochemistry of the Warrumbungle Volcano, New South Wales, PhD Thesis, University of Sydney. 52 Ghorbani, M.R. (2003). Phonolitic and trachytic rocks from the Warrumbungle volcano, different sources and conditions. EGS-AGU Joint Assembly, Nice, France, 6—I1 April 2003, abstract # 9241. Ghorbani, M. and Middlemost, E.A. K. (2000). Geochemistry of pyroxene inclusions from the Warrumbungle Volcano, New South Wales. American Mineralogist 85, 1349-1367. Gibson, D.L. (2007). ‘Potassium-argon ages of Late Mesozoic and Cainozoic Igneous Rocks of Eastern Australia’. CRC LEME Open File Report 193. Gibson, D.L. (2008). ‘Landscape evolution: a component of catchment characteristics’. 2” International Salinity Forum, 31 March-3 April, Adelaide, South Australia. Final Papers [PDF]. Gold, H. and Prineas, P. (1997). ‘Wild Places: Wilderness in Eastern New South Wales. Second edition, revised’. (Coolong Foundation for Wilderness Ltd: Sydney). Gordon, J. and Brown, E. (2010). New approaches — geodiversity and ecosystem services. Earth Heritage 35, 22-23. Hill, L., Peake, T. and Bell, s. (2001). “Vegetation Survey and report on Towarri National Park, Cedar Brush Nature Reserve and Wingen Maid Nature Reserve’. (NWS unpublished report: Sydney). Hill, P., Rollett, N. and Symonds, P. (2001). ‘Seafloor mapping of the South-east Marine Region and adjacent water-AUSTREA final report: Lord Howe Island, south-east Australian margin (includes Tasmania and South Tasman Rise) and central Great Australian Bight’. (AGSO Record 2001/ 08: Canberra). Hutchison, Lancia (2010). “Mists of the Magic Cauldron’. (IB Publications: South Murwillumbah). Hutton, I. (2008). ‘A Guide to World Heritage Lord Howe Island’ (Lord Howe Island Museum: Lord Howe Island). Jenkins, B. and Morand, D. (2002). A comparison of basaltic soils and associated vegetation patterns in contrasting climatic environments. In ‘Regoliths and Landscapes in Eastern Australia’ (Ed. I.C.Roach) pp 26-30. (CRC LEME: Perth). Johnson, D.P. (2004). ‘The Geology of Australia’. (Cambridge University Press: Cambridge). Kariuki, M., Kooyman, R.M., Smith, R.G. B., Wardell- Johnson, G. and Vanclay, J. K. (2006). Regeneration changes in tree species abundance, diversity and structure in logged and unlogged sub-tropical forests over a thirty six year period. Forest Ecology and Management 238 (2-3), 162—176. Kauhikaua, J., Hildenbrand, T. and Webring, M. (2000). Deep magmatic structures of Hawaiian volcanoes, imaged by three-dimensional gravity models. Geology 28, 883-886. Knessel, K.M., Cohen, B.E., Vasconcelos, P.M. and Thiede, D.S. (2008). Rapid change in drift of the Australian plate records collision with Ontong Java plateau. Nature 454, 754-758. Proc. Linn. Soc. N.S.W., 132, 2011 F.L. SUTHERLAND Knutson, J. (1989). Comboyne. In ‘Intraplate Volcanism in Eastern Australia and New Zealand (Compl. R.W. Johnson) pp. 124-125. (Cambridge University Press: Cambridge). Lewis, G.B., Mattox, S.R., Duggan, M. and McGee, K (1998). “Australian volcanoes educational and slide set’. (Australian Geological Survey Organisation: Canberra). MclIntrye-Tamwoy, S. (2008). Archaeological sites and indigenous values: the Gondwana Rainforests of Australia World Heritage Area. Archaeological Heritage 1, 42-49. Manglik, A. and Christensen, U-.R. (2006). Effect of lithospheric root on decompression melting in plume- lithosphere interaction models. Geophysics Journal International 164, 259-270. Mitchell, G. (1990). Ben Hall’s Gap. Land of the Dinosaurs. Habitat Australia 18 (6), 24-27. Mitchell, P.B. (2003). ‘NSW ecosystems, data base mapping unit descriptions. Unpublished Report’. (NSW National Parks and Wildlife Service: Hurstville). Mortimer, N., Gans, P.B., Palin, J.M., Meffre, S., Herzer, R.H. and Skinner, D.N.B. (2010). Location and migration of Miocene-Quarternary volcanic arcs in the SW Pacific region. Journal of Volcanology and Geothermal Research 190, 1-10. Nekvasil, H., Dondolini, A., Horn, J., Filiberto, J., Long, H. and Lindsley, D.H. (2004). The origin and evolution of silica-saturated alkalic suites: an experimental study. Journal of Petrology 45, 693— 720e Nott, J., Young, R. and McDougall, I. (1996). Wearing down, wearing back, and gorge extension in the long term denudation of a highland mass: quantitative evidence from the Shoalhaven Catchment, Southeast Australia. The Journal of Geology 104, 224-232. Och, D.J., Offier, R., Zwingmann, H., Braybrooke, J. and Graham, I.T. (2009). Timing of brittle faulting and thermal events, Sydney region: association with early stages of extension of East Gondwana. Australian Journal of Earth Sciences 56, 873-887. Ollier, C. and Pain, C.F. (2000). ‘The origin of mountains’. (Routledge: London). O’Reilly, S.Y. and Zhang, M. (1995). Geochemical characteristics of lava field basalts from eastern Australia and inferred sources: connections with sub—continental lithospheric mantle. Contributions to Mineralogy and Petrology 121, 148-170. Osborne, R.A.L., Docker, B. and Salem, L. (1998). Places of Geoheritage Signficance in New South Wales Comprehensive Regional Assessment (CRA) Forest Region, Unpublished Report, 106 pp. (Resources and Conservation Division, Department of Urban A ffairs and Planning: Sydney). Pain, C.F. and Ollier, C.D. (1986). The Comboyne and Bulga Plateaus and the evolution of the Great Escarpment. Journal and Proceedings of the Royal Society of New South Wales 119, 123-130. Parfitt, E.A. and Wilson L. (1995). Explore volcanic eruptions IX: The transition between Hawaiian-style Proc. Linn. Soc. N.S.W., 132, 2011 lava fountaining and Strombolian explosive activity. Geophysical Journal International 121, 226-322. Parfitt, E.A. and Wilson L. (1999). A Plinian treatment of fallout from Hawaiian lava fountains. Journal of Volcanology and Geothermal Research 88, 67-75. Pogson, D.J. and Watkins, J.J. (1998). ‘Bathurst 1: 250 000 Geological Sheet (SI/ 55—8): Explanatory Notes’ (Geological Survey of New South Wales: Sydney). Reitsma, J. and Allen, R.M. (2003). “The elusive mantle plume’. Earth and Planetary Science Letters 207, 1-12. Rey, P. F. and Muller, R.D. (2010). Fragmentation of active continental plate margins owing to the buoyancy of the mantle wedge. Nature Geoscience 3, 257-261. Roach, I.C., McQueen, K.G. and Brown, M.C. (1994). Physical and petrological characteristics of basaltic eruptive sites in the Monaro Volcanic Province, southeastern New South Wales, Australia. AGSO Journal of Australian Geology and Geophysics 15, 381-394. Roberts, D.L., Sutherland, F.L., Hollis, J.D., Kennewell, P. and Graham, I.T. (2004). Gemstone characteristics, North—East Barrington Plateau, NSW. Journal and Proceedings of the Royal Society of New South Wales 137, 99-122. Scheibner, E. (1999). “The Geological Evolution of New South Wales- A brief review’. (Mineral Resources New South Wales: Sydney). Schon, R.W. (1984). “The Geological Heritage of New South Wales- Volume III’. (Geological Society of Australia, New South Wales Division: Sydney). Schon, R.W. (1989). Liverpool Range. In ‘Intraplate Volcanism in Eastern Australia and New Zealand’. (Compl. R.W. Johnson) pp. 122—123. ( Cambridge University Press: Cambridge). Sharp, K.R. (2004). Cenozoic volcanism, tectonism and stream derangement in the Snowy Mountains and northern Monaro of New South Wales. Australian Journal of Earth Sciences 51, 67-85. Sheth, H.C. (2003). The active lava flows of Kilauea volcano, Hawaii. Resonance 8 (6), 24-33. Stevens, N. and Willmott, W. (1996). ‘Rocks and Landscape Notes. Main Range’. (Geological Society of Australia, Qld Divison: Brisbane). Stevens, N. and Willmott, W. (1998). “Rocks and Landscape Notes. Mount Barney-Mount Ballow’. (Geological Society of Australia, Qld Division: Brisbane). Sutherland, F.L. (1995). ‘The Volcanic Earth’. (UNSW Press: Sydney). Sutherland, F.L. (2003). ‘Boomerang’ migratory intraplate Cenozoic volcanism, eastern Australian rift margins and the Indian-Pacific mantle boundary. Geological Society of Australia Special Publication 22 and Geological Society of America Special Paper 372, 203-221. Sutherland, F.L. and Barron, L.M. (2003). Diamonds of multiple origins from New South Wales: further data and discussion. Australian Journal of Earth Sciences 50, 975-981. 53 GEODIVERSITY AND HABITAT IN VOLCANIC AREAS Sutherland, F.L. and Fanning, C.M. (2001). Gem-bearing basaltic volcanism, Barrington, New South Wales: Cenozoic evolution based on basalt K-Ar ages and zircon fission track and U-Pb isotope dating. Australian Journal of Earth Sciences 48, 22\—237. Sutherland, F.L., Colchester, D.M. and Webb, G.B. (2005a). An apparent diatreme source for gem corundum and zircon, Gloucester River, New South Wales. Journal and Proceedings of the Royal Society of New South Wales 138, 77-84. Sutherland, F.L., Graham, I.T. and Zwingmann, H., Pogson, R.E. and Barron, B.J. (2005b). Belmore volcanic province, northeastern New South Wales and some implications for plume variations along Cenozoic migratory trails. Australian Journal of Earth Sciences 52, 897-919. Sutherland, F.L., Graham, I.T., Pogson, R.E., Schwarz, D., Webb, G.B., Coenraads, R.R., Fanning, C.M., Hollis, J.D. and Allen, T.C. (2002). The Tumbarumba basaltic gem field, New South Wales, in relation to sapphire-ruby deposits of eastern Australia. Records of the Australian Museum 54, 215-248. Sutherland, L. and Graham, I. (2003). ‘Geology of the Barrington Tops Plateau’. (The Australian Museum Society: Sydney). Sutherland, L. jnr (2008). ‘Crater of Life [DVD]’. (Below H,0 Productions: Currumbin, Qld). Tacon, P.S.C. (1998). Identifying Ancient Sacred Landscapes in Australia: From Physical to Social. In ‘Archaeologies of Landscape: Contemporary Perspectives’ (Eds Ashmore W. and Knapp A.B.) pp 35-57. (Blackwell Publishers: Oxford). Taylor, G. and Roach, I.C. (2003). Monaro region, New South Wales. Unpublished Report, 6pp. (CRC LEME: Perth, WA). Taylor, G., Trusswell, E.M., McQueen, K.G. and Brown, M.C. (1990). Early Tertiary palaeogeography, landform evolution and palaeoclimates of the southern Monaro, N.S.W., Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 78, 109-134. Thompson, D., Bliss, P. and Priest, J. (1987). “Lord Howe Island Geology’. (Geological Survey of New South Wales: Sydney). Thompson, J. (1974). “Drake 1:100 000 Geological Sheet 9340, 1 Edition’. (Geological Survey of New South Wales: Sydney). Tomkins, K.M. and Hesse, P.P. (2004). Evidence of Late Cenozoic uplift and climate change in the stratigraphy of the Macquarie River valley, New South Wales. Australian Journal of Earth Sciences 51, 273-290. UNESCO (2010). ‘The World’s Heritage’. (UNESCO Publishing: Paris and Harper Collins: London) Uni Serve-Science (2001). ‘Oz Volcanoes’. (University of Sydney: Sydney). United Nations Environment Program-Wo (2008). Central Eastern Rainforest Reserves, Australia. In “Encyclopedia of the Earth’. (Ed. C.J. Cleveland). (Environment Information Coalition, National 54 Council for Science and the Enviornment: Washington DC). Van der Beek, P., Pulford, A. and Braun, J. (2001). Cenozoic landscape development in the Blue Mountains (SE Australia): lithological and tectonic controls on rifted margin morphology. Journal of Geology 109, 35—S6. Van der Zander, I., Sinton, J.M. and Mahoney, J.J. (2010). Late shield stage silicic magmatism at Waianae Volcano: Evidence for hydrous crustal melting in Hawaiian volcanoes. Journal of Petrology 51, 671-701. Vasconcelos, P.M., Knessel, K.M., Cohen, B.E. and Heim, J.A. (2008). Geochronology of the Australian Cenozoic — a history of tectonic and igneous activity weathering, erosion and sedimentation. Australian Journal of Earth Sciences 55, 865-914. Veevers, J.J. (2001). ‘ATLAS of Billion-year earth history of Australia and neighbours in Gondwanaland’. (GEMOC Press: Sydney). Vickery, N.M., Dawson, M.W., Sivell, W.J., Malloch, K.R., and Dunlap, W.J. (2007). Cainozoic igneous rocks in the Bingara to Inverell area, northeastern New South Wales. Quarterly Notes of the Geological Survey of New South Wales 123. Warth, D. (2009). ‘Rainforest [DVD]. The Secret of Life’. (David Warth Productions, Byron Bay, NSW). Westerman, H. (2004). ‘Geological information on Three Nature Reserves in the Lansdowne Area Unpublished Report’. (NSW National Parks and Wildlife Service, Manning area: Taree). Whitehead, J. (2008). ‘The Warrumbungles: dead volcanoes, national parks, telescopes and scrub’. (Coonabarabran, NSW). Whitehead, J. (2009). ‘The Warrumbungle volcano: a geological guide to Warrumbungle National Park’. (Warrumbungle National Park, Conabaraban). Willmott, W.F. (2003). “Rocks and Landscapes of the National Parks of Southern Queensland’. (Geological Society of Australia, Queensland Division: Brisbane). Willmott, W.F. (2010). “Rocks and Landscapes of the Gold Coast Hinterland’. Expanded Third Edition. (Geological Society of Australia, Queensland Division: Brisbane). Woodroffe, C.D., Brooke, B.P., Linklater, M., Kennedy, D.M., Jones, B.G., Buchnan, C., Mleczko, R.., Hua, Q. and Zhao, J-X (2010). Response of coral reefs to climate change: Expansion and demise of the southernmost Pacific coral reef. Geophysical Research Letters 37, L15602, 6pp. Wright, P. (1996). “The NPA Guide to National Parks of Southern New South Wales’. National Parks Association of NSW: Sydney). Young, R.W. and McDougall, I. (2004). Cenozoic volcanism, tectonism and stream derangement in the Snowy Mountains and northern Monaro of New South Wales. Australian Journal of Earth Sciences 51, 765-772. Zoete, T. (2000). Vegetation survey of the Barrington Tops and Mount Royal National Parks for use in Fire Management. Cunninghamia 6, 511-539. Proc. Linn. Soc. N.S.W., 132, 2011 Geodiversity of the Southern Barrington Tops Lava Field, New South Wales: A Study in Petrology and Geochemistry M.C. BRUCE Department of Industry and Investment NSW — Mineral Resources, WB Clarke Geoscience Centre, Londonderry 2753, Australia. Bruce, M.C. (2011), Geodiversity of the southern Barrington Tops Lava Field, New South Wales: A study in petrology and geochemistry. Proceedings of the Linnean Society of New South Wales 132, 55-69. The Barrington Tops lava field lies within the Barrington Tops National Park and State Forests north of the Hunter River Valley and west of Gloucester in eastern NSW. Mapping in the southern lava field has identified 33 basaltic flows each 10-20m thick and separated by agglomerate or by sub-horizontal palaeosols. Petrography and whole-rock geochemistry indicate basanites dominate the stratigraphic sequence, with subsidiary alkali basalts, pyroxene-phyric basalts (ankaramites) and tholetites. Alkali gabbros form intrusions high in the sequence and represent conduits for surface lavas. Modelling of major and trace elements from the alkaline rocks reveal a co-genetic relationship, chiefly controlled by a low pressure olivine + plagioclase mineral assemblage. Incompatible trace elements suggest low degree melting of an enriched mantle source with entrained amphibole-enriched sub-continental lithospheric mantle. A more depleted mantle component may have contributed to the tholeiites. These southern findings are consistent with observations in the northern lava field, and help to model evolution of the Barrington Tops volcano. Topographic inversion has contributed to geomorphological features which support diverse floral and faunal communities. Such diversity is underpinned by the geology and in particular the lava field, which shaped the natural landscape, provides fertile soils and releases gem-quality sapphires, zircons and rubies. Manuscript received 1 November 2010, accepted for publication 20 April 2011. KEYWORDS: Barrington Tops, basalt, geochemistry, geodiversity, lava field, New South Wales, petrology, INTRODUCTION Barrington Tops is located approximately 100km north-northwest of Newcastle, New South Wales (Fig. 1; inset). The region is dominated by a plateau of average elevation approximately 1500m above sea level with a relief of over 1100m to the valley floors. Barrington Tops is rugged, heavily vegetated and incorporates the Barrington Tops National Park and several surrounding State Forests. Both the National Park and State Forests are well known for recreational purposes that derive from the environmental diversity and geomorphology — both dependent on the underlying geology — of this remarkable region. The Barrington Tops lava field has a K-Ar age of 59 to 44 Ma (Wellman et al. 1969; Sutherland and Fanning 2001), although more recent fission track and U-Pb dating of zircons suggests that parts of the field were active for over 55 million years, until 4 Ma (Roberts et al. 2004). The volcanic field consists predominately of alkali basaltic rocks, with some olivine tholeiites at the base of the sequence (Mason 1982). Following Mason’s (1982) interpretation of the basalts as flows, Pain (1983) envisaged that they were derived from a shield volcano, an idea supported by subsequent authors (O’ Reilly and Zhang 1995; Sutherland and Fanning 2001; Sutherland and Graham 2003). The Cenozoic lavas in part overlie Late Permian granitoids which were emplaced at high crustal levels (3-7 km: Eggins 1984) and partly within low grade metamorphosed Carboniferous and Devonian sediments (Mason and Kavalieris 1984) of the New England Orogen that the granites intrude. In this paper, I present data (collected in 1995 during my Honours thesis) from the southern part of the volcanic field (Fig. 1), an important addition to the literature as most data for these Paleocene/ Eocene (51 to 59 Ma) basalts come from the northern side (Wellman et al. 1969; Mason 1982; Pain 1983; O’Reilly and Zhang 1995; Sutherland and Fanning 2001). The southern lavas range from SiOz -deficient basanites, akali basalts and rare trachybasalts to SiOz SOUTHERN BARRINGTON TOPS LAVA FIELD 32°00'S 32°05’S Sat Tat et 4.4.4.9 FP 2 Fs <.¢.<¢ 4 | _] Metasediments 151225 Granodiorite Basaltic lava field Kilometres Figure 1. Simplified geological map of the southern Barrington Tops lava field. Numbered points refer to localities mentioned in the text: 1, Williams Range (Careys Peak Trail); 2, Allyn Range; 3, Mount Royal Range; 4, Mt Allyn; 5, Mt Lumeah; 6, Careys Peak; 7, Mt Barrington; 8, Barrington Falls. Inset indicates the map area within NSW. saturated tholeiites, interspersed with pyroxene- phyric flows (ankaramites) and alkali gabbros. STRATIGRAPHY Local stratigraphy in the southern Barrington Tops basaltic field has been determined from detailed petrological study of individual lava flows from type sections exposed in several localities along the Careys Peak Trail, Mounts Allyn, and Lumeah and the Barrington Tops plateau (Fig. 1). Correlations between sites were carried out topographically 56 because of the sub-horizontal nature of the basalt anda similar topographic height of the basalt-metasediment contact. This contact occurs at approximately 920m above sea level. Fig. 2 is a stratigraphic sequence of the lavas differentiated on petrological and chemical criteria rather than on individual flows. The base of the sequence consists of a thin layer of alkali basalts (~20m thick) and tholeiites (~110m thick), with basanites dominating the middle section (> 200m thick). Interrupting the continuity of the basanites is a relatively thin succession of trachybasalts (~30 m thick). The upper part of the sequence comprises Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE 1555m 1500m 1440m 1390m 1370m 4250m Cenozoic 1220m 1140m 1080m 940m 920m Carboniferous Alkali basalt and gabbro Pyroxene-phyric basalt Alkali basalt Basanite Alkali basalt Basanite Trachybasalt Basanite Untyped basalt Tholeiite Alkali basalt Pyroxene-phyric basalt Alkali basalt Undifferentiated beds Figure 2. Volcanic stratigraphy for the southern part of the lava field. Delineations are based on changes in rock-type rather then individual flows. In all 33 flows are exposed through 630m of topographic relief. Numbers refer to elevation above sea level. alternating successions of alkali basalt and basanite (~185m thick). Pyroxene-phyric basalt (ankaramite) occurs as thin layers both at the base and the top of the sequence in close association with the alkali basalts. In all 33 individual flows are found to make up the sequence exposed through 630m of relief. Most volcanic flows are identified by the presence of an agglomerate/breccia up to 3 metres thick, which — given its regular occurrence every 10 to 20 metres in the succession — is interpreted to represent either the top or base of individual flows. Some volcanic flows, however, are conspicuous by the absence of the agglomerate. These flows are usually the most vesicular (due to passively expelled volatiles) and are commonly separated by sub-horizontal palaeosols, indicating a temporary cessation of volcanism with Proc. Linn. Soc. N.S.W., 132, 2011 weathering and erosion. Alkali gabbro (teschenite) is also present on the Barrington Tops plateau at Careys Peak and to the south of Mount Barrington above an apparent magma chamber (Wellman 1989). The alkali gabbro was originally noted by Benson (1912) and interpreted as intrusive dykes or sills. Measurement of magnetic anisotropy from one basanite (Barrington Falls) and one tholetite (Williams Range) indicates that the dominant lineation in these samples strikes approximately north-south and northwest-southeast respectively. Projection of these directional lines intersects on the alkali gabbro to the south of Mount Barrington (Bruce 1995), suggesting that this intrusive body represents remnant vents from which the lava was extruded. 57 SOUTHERN BARRINGTON TOPS LAVA FIELD PETROGRAPHY AND MINERAL CHEMISTRY Analytical Techniques Minerals were analysed for major elements using a Cameca SX50 Electron Microprobe calibrated with natural and synthetic materials with a general precision < 1%. Analytical conditions were optimised for a standard silicate run using a 15kV accelerating elements, with the exception of K and Na for which a broader (10nA) beam was used. Routine analyses were obtained by counting 30s at peak and 5s on background. Alkali Basalt (Fig. 3a) The alkali basalts are fine-grained in hand specimen with scattered phenocrysts of greenish olivine up to 5mm in size. In thin section, the fine- voltage and a 20nA focussed electron beam for all Fig. 3, (a). Photomicrograph (cross polarisation) of an alkali basalt illustrating olivine phenocrysts scattered throughout the groundmass with abundant plagioclase laths. Magnification x 2.5. (b). Pho- tomicrograph (cross polarisation) of a tholeiite displaying ophitic texture defined by plagioclase laths embedded in clinopyroxene. Magnification x 2.5. (c). and (d). Photomicrographs (plane polarisation) from basanites demonstrating three different clinopyroxene textures. Top photo illustrates a sieve-tex- tured proxene (top centre). Bottom photo shows pyroxene crystals in a radiating texture (left centre) and small euhedral phenocrysts (pinkish tinge). Magnification x 6.3. (e). Oscillatory zoning in a twinned Ti-rich diopside phenocryst from a pyroxene-phyric basalt. Cross polarisation. Magnification x 6.3. (f). Photomicrograph of a holocrystalline alkali gabbro dominated by plagioclase laths, olivine and clinopy- roxene crystals. Cross polarisation. 58 Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE grained groundmass is composed of intergranular olivine, clinopyroxene, plagioclase, Fe-Ti oxides and intersertal glass. Subhedral phenocrysts and microphenocrysts of olivine are present (0.1-2mm) as are abundant calcic plagioclase laths (0.2mm). The larger olivine phenocrysts are commonly zoned from a relatively Mg-rich core (Fo7z) to more Fe-rich rims (Foes). Clinopyroxene is notable in its absence as a phenocryst phase and is restricted to the groundmass where it is of diopside composition (En,,Fs,,Wo,,). Zeolites are also present as a minor secondary phase and occur as amygdules (0.2mm). Tholeiite (Fig. 3b) The tholeiites are very fine-grained and dark grey in hand specimen. In outcrop they are horizontally (?flow) layered. In thin section they are ophitic to subophitic in texture and consist of clinopyroxene and olivine phenocrysts in a groundmass of plagioclase laths, clinopyroxene and Fe-Ti oxides. Rare, prism- shaped plagioclase phenocrysts (~0.4mm) are also present. The plagioclase laths with an average length of 0.1mm are embedded in augite crystals which » average 0.7mm in size. Olivine microphenocrysts, however, do not contain any plagioclase inclusions. Intersertal glass and chlorite is evident in the groundmass. Basanite (Figs 3c and 3d) The basanites are very fine-grained in hand specimen with small phenocrysts (medium 0.5mm) of olivine and clinopyroxene. In thin section, olivine (Fosa-90) is the most abundant phenocryst, averaging 0.5mm with a maximum of 2mm. Olivine microphenocrysts (0.2mm) and similar sized calcic plagioclase laths (Aneo-71) are dominant in a fine- grained groundmass of intergranular and interstitial clinopyroxene, olivine, magnetite, calcic plagioclase and mesostasis feldspar. Zeolites usually occur as small amygdules but are also observed as large crystals up to 2mm in size. Glomeroporphrytic aggregates of olivine are common in some samples. Three types of clinopyroxene phenocrysts occur within the basanites: 1. euhedral pyroxene phenocrysts (average 0.8mm) in which the cores are sieve-textured consisting of probable glass and clinopyroxene, which could not satisfactorily be analysed, and rims that are glass free and composed of diopside (En,,Fs,,Wo,,); 2. small (0.3mm) radiating crystals that are zoned from a diopside core (En,,Fs,,Wo,,) to a Ti-rich aluminium diopside rim (En,,Fs,,Wo,,); 3. large (up to 0.9mm) subhedral to euhedral clinopyroxene phenocrysts zoned from cores of En,,Fs,,Wo,, to rims of Ti-rich aluminium diopside (En,,Fs,,Wo,,). The first type of Proc. Linn. Soc. N.S.W., 132, 2011 phenocryst is interpreted to represent slow growth of clinopyroxene in which the core has trapped melt before crystallising the outer margin probably during quenching. The second type represents a snapshot of the early stages of pyroxene growth. Further growth of this type has resulted in the third type. In addition, clinopyroxene xenocrysts are evident in some samples. These are commonly large in size (up to 9mm) and sub-rounded in shape. A gabbroic enclave (Smm) that occurs in one sample consists of clinopyroxene (En,Fs,,Wo47), orthopyroxene (En,,Fs,,) and calcic plagioclase (An,Ab,,). The presence of orthopyroxene in apparent equilibrium with clinopyroxene suggests that the enclave was formed under high pressure and transported to the surface rapidly to prevent resorption during ascent (IT. Green pers. comm. 1995). This is supported by aluminium stoichiometry in the clinopyroxenes where Al*:Al® = 1.8, suggesting crystallisation pressures >1 GPa (Thompson 1974; Wass 1979) or >35 km in depth. However, the relatively low AlzOs content of the co-existing orthopyroxenes (2.2 wt%) is not consistent with an upper mantle origin (Binns et al. 1970) and therefore probably crystallised at crustal pressures. Pyroxene-phyric Basalt (Fig. 3e) These basalts (ankaramites) are conspicuous in hand specimen due to an estimated 30-40 percent of large (up to 10mm) clinopyroxene megacrysts. In thin section, the megacrysts are diopside in composition and display strong oscillatory zoning with Mg-rich cores (En,,Fs,,Wo,,) and Fe-Ca-rich rims (En,,Fs,,Wo,,). They are also Ti-rich with 2-3 wt% TiOz. The megacrysts contain various inclusions of diopside (En,,Fs,,Wo,,), olivine (Foz), plagioclase (AnecsA baa) and Ti-rich magnetite. Olivine phenocrysts (1-2mm) zoned from cores of Fo74 to rims of Foes and plagioclase laths (AnesAbsa) up to 2mm, are abundant and embedded in a fine hypocrystalline groundmass of clinopyroxene (EnssFsisWoa7), olivine (Foez2), calcic plagioclase and magnetite - ulvéspinel. Both zoning in the phenocrysts and megacrysts and inclusions in the megacrysts would suggest variable crystal growth rates, typical of cumulative textures. Alkali Gabbro (Fig. 3f) The alkali gabbros are holocrystalline and consist of Ti-rich diopsides up to 5mm in size and 2mm olivine phenocrysts. The diopsides are zoned from En,,Fs,,Wo,, to En,Fs,Wo,, with entrained plagioclase (Anez) and magnetite - ulvdéspinel inclusions. Olivine phenocrysts commonly form a 59 SOUTHERN BARRINGTON TOPS LAVA FIELD glomeroporphrytic texture and can be strongly zoned from Fo71 cores to more Fe-rich (Foss) rims. Large plagioclase (An,,) laths up to 3mm are abundant and set in a coarse (>0.5mm) groundmass of clinopyroxene (En,.Fs,,Wo,.,), forsterite, mesostasis feldspar, magnetite - ulv6spinel, analcime and apatite needles. WHOLE-ROCK COMPOSITIONS Analytical Techniques Samples were crushed in a TEMA tungsten carbide mill. Major elements were determined using glass fusion discs (Norrish and Hutton 1969) by X-ray Fluorescence (XRF). The instrument used was a Siemen’s SRS-1. Calibration was by means of international rock standards and well-calibrated internal standards. All samples were run in duplicate. Trace elements were analysed by XRF using pressed- powder pellets. Mass absorption corrections were applied (Norrish and Chappell 1977). All samples were analysed in duplicate, using international rock standards. FeO was determined by HF digestion and titration with Ceric sulphate. Estimates of precision based on USGS standards BCR-1 and GSP-1, and NIM standard NIM-G are: Major elements; < +1% relative at > 10 wt% levels; Mn, P, Mg, Na < 2% relative for BCR-1: Trace elements; < 1% relative at 100ppm levels; + 10ppm for < 100ppm. Results Selected bulk-rock data representing each described rock-type are presented in Table 1. The results for all samples analysed are tabulated in the Appendix (Table A1). Major Elements The majority of sampled _ basalts are undersaturated basanites and alkali basalts according to the classification of Le Bas et al. (1986), although one tholeiite and one trachybasalt has also been sampled (Fig. 4). The pyroxene-phyric basalt and the alkali gabbro when considered as plutonites plot as monzogabbro and gabbro respectively using the scheme of Middlemost (1985; not shown). Major elements versus Mg# [100Mg/(Mg + Fe **)] are plotted in Fig. 5 for all samples except the tholeiite. Although some diagrams show substantial scattering, both AlzOs and MgO display strong, coherent trends. This would suggest that all alkaline magmas are related petrogenetically via processes dominated by compositions that concentrate the elements Al2O3 and MgO (fractionation/accumulation, partial melting, 60 mixing). Overall there is an increase in TiOz, AlzOs, FeOt (total iron) and K2O and a decrease in MgO and CaO with increasing fractionation (decrease in Mg#). A positive correlation between CaO/Alz2O3 and Mg# for the basanites and alkali basalts with Mg# <70 would indicate clinopyroxene fractionation in these rocks. Trace Elements Plots of the compatible elements Ni and Cr versus Mg# (Fig. 6) define a strong, positive correlation for the former but a more scattered relationship for the latter. This, combined with a similar MgO trend, suggests that olivine fractionation was the dominant process by which the alkaline magmas were related. Clinopyroxene may not have been a significant fractionating phase in controlling magma composition in the basanites and alkali basalts given the reasonably constant Cr concentrations; although it could have been important in the evolution of the trachybasalts, pyroxene-phyric basalts and alkali gabbros. Zr/Nb values are virtually constant within a magmatic suite as crystal fractionation and wall rock reaction have little effect on this ratio (Green 1992). This ratio does, however, increase with increasing degrees of partial melting, from values around 3-6 for oceanic island basalts (OIB) to > 30 for normal mid ocean ridge basalts (N-MORB) (Crawford et al. 1997). The alkaline magmas in the southern Barrington Tops lava field have Zr/Nb = 2-3, whereas the sampled tholeiite has a Zr/Nb value > 4. The higher ratio in the tholeiite could indicate either a different, or a heterogeneous, magma source. Furthermore, the source(s) for both magma lineages have Zr/Nb ratios typical of OIB. Multi-element diagrams for both the alkaline magmatic suite and the tholeiite are presented in Fig. 7. These plots are normalised to the average OIB values of Sun and McDonough (1989). Overall, the alkaline magmas have patterns and element concentrations similar to OIB. However, relative to the average OIB, the rocks are depleted in Rb, K, Zr, Ti and Y and enriched in Ba, Nb, Sr and P. The tholeiite follows a similar pattern but has concentrations more akin to enriched mid ocean ridge basalt (E-MORB) suggesting a slightly more depleted component in its source. PETROGENESIS Major and trace elements indicate that the alkaline magmas were strongly controlled by the fractionation of a phase dominated by MgO and Ni content and accumulation of a phase dominated by Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE Table 1 Selected major and trace element XRF analyses, southern Barrington lava field. Sample MU55361 MU55372 MU55418 MU55365 MU55369 MU55397 Barrington Tops 4:25k 9133-1-N GR.492538 GR.553522 GR.560480 GR.500539 GR.502533 GR.555490 Pyroxene-phyric Rock type Alkali basalt Basanite Tholeiite basalt Alkali gabbro Trachybasalt wt% SiO2 47.44 44.16 49.15 45.25 48.9 46.1 TiO; 1.97 2.53 1.49 2.55 2.3 2.65 Alz03 14.49 14.54 16.42 16.01 16.38 15.97 Fe203 1.93 1.91 2.09 1.85 1.85 1.9 FeO 9.63 9.55 10.44 9.28 9.23 9.49 MnO 0.17 0.19 0.16 0.18 0.16 0.18 MgO 10.44 10.95 6.33 7.31 6.28 7.99 CaO 9.8 11.5 10.03 11.73 9.67 9.19 Na,O 2.72 2.98 2.77 3.32 3.28 3.32 K,0 1.11 0.85 0.52 1.77 1.24 1.78 P205 0.39 1.07 0.32 0.96 0.5 0.88 Total 100.09 100.23 99.74 100.21 99.79 99.45 CIPW norms Or 6.56 5.02 3.08 10.45 7.35 10.59 Ab 20.75 11.24 23.48 9.5 27.78 18.8 An 23.99 23.7 30.88 23.46 26.32 23.5 Ne 1.2 7.53 10.03 0 5.1 Di 18.06 21.75 14.17 23.64 15.42 13.96 Hy 0 0 15.28 0 1 0 Ol 22.06 20.88 6.54 13.33 13.96 18.29 Mt 2.8 2.76 3.04 2.68 2.69 2.77 ll 3.74 4.8 2.84 4.83 4.38 5.06 Ap 0.85 2.33 0.7 2.09 1.09 1.93 An/An+Ab 0.536 0.678 0.568 0.712 0.487 0.556 Mg/Mg+Fe** 0.659 0.672 0.519 0.584 0.548 0.600 ppm Ba 409 580 146 712 381 606 Rb 20 22 33 17 22 Sr 634 1146 412 1017 666 1328 Y 20 23 22 27 25 27 Zr 122 246 80 226 158 324 Nb 46 103 19 115 53 121 Th 1 6 0 7 1 3 Pb 5 3 3 3 5 2 Ga 19 20 21 23 24 19 Zn 90 80 80 76 84 83 Cu 43 51 71 62 43 33 Ni 171 177 208 82 52 107 V 228 243 192 240 219 195 Cr 351 337 294 112 84 115 AlzOs. In a basaltic system, the two phases likely to be involved are olivine and plagioclase respectively. However, Harker-style plots cannot preclude the importance of clinopyroxene as a fractionating phase. Therefore the mineral assemblages; olivine + plagioclase and olivine + plagioclase + clinopyroxene (amongst others) are tested below using the model of Pearce (1968). In this model, element ratios are Proc. Linn. Soc. N.S.W., 132, 2011 used instead of oxide-oxide wt% diagrams as the latter may produce spurious correlations (Russell and Nicholls 1988; Rollinson 1993). These ratios are referred to as Pearce Element Ratios (PER) and are based on stoichiometry of the end-members of the mineral phases. The data almost always falls on a straight line (within analytical errors) and by applying a least squares linear regression technique, the slope SOUTHERN BARRINGTON TOPS LAVA FIELD Ultrabasic Alkaline Foidite ‘ ! Basaltic ! Tephtite trdchy- Basahite Na,O+K20 basalt Basaltic andesite Picrobasalt 50 Intermediate Phonolite ' Tradhyte Trachydacite ' Rhyolite ! \ ! ! ! Ibe “11 ! ! ! I Andesite Subalkaline/Tholeiitic 60 70 80 SiO, Figure 4. Rock classification diagram of Le Bas et al. (1986). Plotted are the various basaltic lithologies from the southern part of the lava field. Symbols: © = basanite; 9 = alkali basalt; x = trachybasalt; + = pyroxene-phyric basalt; ¢ = alkali gabbro; v = tholeiite. of the line can be calculated. This slope reflects the stoichiometry of the minerals, which should equal one (20 error) if the minerals tested for are in fact the phases involved in the differentiation of the magma. Fig. 8 presents PER diagrams for the following mineral assemblages within a basaltic system; olivine, clinopyroxene, plagioclase, olivine + clinopyroxene, olivine + plagioclase, olivine + clinopyroxene + plagioclase. The pyroxene-phyric basalt and the alkali gabbro have been removed from the model because of the abundance of large crystals that are highly likely to be cumulates from the magmas. The mineral assemblage olivine + plagioclase is the only assemblage that produces a slope of one (0.97) within the 95% confidence limit. Thus, the PER model validates the linear trends of MgO and AlzOz observed in the Harker-style diagrams as fractionation/accumulation controlled. | Despite 62 the dominant assemblage of olivine + plagioclase, clinopyroxene must still have been involved in the fractionation process as phenocrysts of this mineral do occur in the basanites, trachybasalts, ankaramites and alkali gabbros. The mantle reservoir for the alkaline magmas has been established as an OIB-type source. Greater enrichment in Nb, Sr and P coupled with depletions in Ti and Y than the average OIB can be explained by smaller degrees of partial melting of a fertile mantle source. However, simple melting models alone cannot explain the lower than expected depletions of Rb, K and Zr. O’Reilly and Zhang (1995) attributed similar Rb and K depletions from the western Barrington Tops lava field to the melting of metasomatised mantle in which residual amphibole was retained in the source. This explanation is also appropriate to the southern part of the lava field. The low Zr anomalies Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE 15.5 16.5 NEE 20S 2:2 24s 26 14.5 1.6 1325 12.0 12 10 11.0 10.0 9.5 11.2 11.6 12.0 06 08 10 12 14 16 1.8 10.8 Fig.5 Major elements (wt%) versus Mg# {100Mg/(Mg+Fe?')}. Same symbols used as in Figure 4. Nb. Tholeiite is not plotted. Proc. Linn. Soc. N.S.W., 132, 2011 63 SOUTHERN BARRINGTON TOPS LAVA FIELD 200 300 400 100 300 400 500 200 100 55 60 65 70 Mg# Figure 6 The compatible elements Ni and Cr (ppm) versus Mg#. Note that the alkali basalt with Ni > 300ppm and Cr > 500ppm has petrographic features indicative of cumulative olivine. Same symbols used as in Figure 4. Nb. Tholeiite is not plotted. could be due to plagioclase accumulation in these rocks as Zr is incompatible in this phase with respect to basaltic melts (Kd = 0.048; Rollinson 1993). Thus, the presence of excess plagioclase in the lavas may have diluted the bulk-rock Zr concentrations, hence the depletions. TOWARDS A GENETIC MODEL Mantle upwelling resulting in lithospheric melting via adiabatic decompression (eg. Hoernle et.al. 2006) is generally accepted as a likely mechanism for the generation of basaltic magmas in continental settings. 64 10 — — — —— — & Basanites Alkali Basalts 10 | ra , 1 x * hig ime m~ % > * * e Trachybasalt > Pyroxene-phyric basalt * Alkali Gabbro 10 E- = =F ae : Tholeiite 15 e 1 ae Rb Th Nb La Sr P Zr Ti Ba K Ta Ce Nd Sm Hf Y Figure 7 OIB-normalised multi-element plots of the various basaltic lithologies from the southern lava field. Normalisation values after Sun and Mc- Donough (1989). Yb Such a scenario (Fig. 9) is envisaged for the southern Barrington Tops lava field, where once formed, some of the magma ascended directly to the surface as evidenced by high-pressure (>1Gpa) gabbroic enclaves and reported mantle xenoliths (Powell and O’Reilly 2007). The majority of the magma, however, was stored in high crustal-level reservoirs conducive to the crystal fractionation of olivine + plagioclase, an assemblage characteristic of low pressure crystallisation. An internal build-up of water in the Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE 20000 18000 o 46000 Slope = 0.30 14000 | = 12000 u% 10000 £ 3000 ix 8000 S 4000 2000 | Ol OUAZ). TEU nl wereeeA Ct TL 0 10000 20000 30000 40000 50000 60000 70000 (Ke SS sss 14000 | Slope = 0.16 : . 12000 aan a £10000 Ses . = 8000 2 : * 6000 & 4000 : 2000 Cpx 0 0 10000 20000 30000 40000 50000 60000 70000 40000 35000 | Slope = 0.51 - 30000 = + LJ 4 oF 0 10000 20000 30000 40000 50000 60000 70000 Si/Zr 45000 mm 40000 i = 35000 Slope = 0.61 % 30000 + 25000 4% 20000 = 15000 Z % 410000 —) ~ «+5000 Ol + Cpx 0 —— ee ooo Q 10000 20000 30000 40000 50000 60000 70000 70000. orm nn nna nnn tn RRNA RRR REA EERE CLC OORE RTA R RETR TT OOO AO CANAL ROE RCE N a = Slope = 0.97 = Li] r4 NN pe .) he + io: = % Ss 0 10000 20000 30000 40000 50000 60000 70000 & 70000 <= = 60000 = 50000 F 4 | ~™ 40000 a | *% 30000 9 x * 20000 in 2 10000 Ol + Cpx + Plag & 0 ——— = 0 10000 20000 30000 40000 50000 60000 70000 S Si/Zr Figure 8. Pearce element ratio diagrams testing for various fractionation/accumulation controlled basaltic mineral assemblages. Note that Ol = Olivine; Cpx = Clinopyroxene; Plag = Plagioclase. Same symbols used as in Figure 4 system ‘pushed’ the cotectic into the clinopyroxene field. The late crystallisation of clinopyroxene may account for the observed phenocrysts in the rocks while obscuring it in the fractionation models. The alkali basalts, whilst free of clinopyroxene phenocrysts, also fractionated the mineral but in this case the pyroxene phenocrysts dropped completely out of the magma and accumulated on the floors and walls of the magma chambers where they continued to grow. Mason (1985) concluded in his study from the northern part of the field that the cores of these phenocrysts possibly crystallised as deep as the core/ mantle boundary based on aluminium stoichiometry. Although similar polybaric crystallisation of pyroxene from ankaramites of the southern lava field (Bruce 1995) suggests shallower depths than this, it certainly does not preclude the existence of deeper level magma chambers, possibly feeding the upper crustal reservoirs. Eruption of these cumulates resulted in the proxene-phyric basalts, which in the field are stratigraphically associated with the alkali basalts. Stratigraphic relationships, magnetic anisotropy Proc. Linn. Soc. N.S.W., 132, 2011 and chemical data are consistent with the genesis of the alkali gabbro as remnant dykes/plugs which necessitated the movement of the magma towards the surface. GEODIVERSITY The southern area of the Barrington Tops lava field has a topographic relief of about 1500m a.s.l whereas the floor of the incised valleys of the Allyn and Williams Rivers has an average elevation of 430m a.s.l. An escarpment marks the southern edge of the plateau. South of the plateau the region is dominated by the Mount Royal, Allyn and Williams Ranges. These ranges are partially capped with basalt while the valleys separating them are not and consist of folded Carboniferous sediments which form the bedrock for the Patterson, Allyn and Williams Rivers. Within these sedimentary units is a conglomerate that crops out along the Allyn River comprised of rounded pebble to boulder-sized granitoids, diorite, siltstones 65 SOUTHERN BARRINGTON TOPS LAVA FIELD Ol+PlagtCpx \ < ~ . - ‘ . . . S ~ S i 7 , 5 Lithosphere Asthenosphere Molten zone \L/ Ol+Plag+Cpx v + Crust ip) Mantle II Mantle Upwelling _ OIB SOURCE RESERVO pte ee ling IRS _ Figure 9. Schematic diagram illustrating the petrogenesis of the basaltic lavas. See text for details. and quartzites cemented in a sandy matrix. Of particular note is the occurrence of an S-type granite clast similar in composition to parts of the Bathurst Batholith (S. Shaw pers. comm.). Rb/Sr dating on muscovite has returned an age of 325 + 3.2 Ma (Bruce 1995), suggesting it may have been sourced from ~300 km to the south-west. Basalt is present from the highest peak (Mount Barrington) down to an altitude of ~920m a.s.l. The resultant volcanic pile is thus ~630m thick. The basaltic flows are sub-horizontal based on the nature of palaeosols which would suggest that the pre-basalt surface consisted of a gently sloping topography. Pain (1983) envisaged a shield volcano for the lava field, an interpretation favoured by most authors (O’Reilly and Zhang 1995; Sutherland and Fanning 2001; Sutherland and Graham 2003). Nevertheless, there is evidence of some localised relief for the pre- basalt surface. Alignment of ferromagnetic minerals, defining lineations or foliations, in individual samples have dips of 20 degrees or more (Bruce 1995). A 66 feature of the basalt today is that it is seen to cover the plateau and ridge tops and is absent in the deeply dissected valleys. This is the opposite of what would be expected had the flows formed on the current topography. It implies that subsequent to the eruption of the basalt, which filled up gently sloping valleys leaving ridges untouched, topographic inversion may have occurred due to a high rate of sedimentary erosion. The previously uncovered ridge tops, consisting of softer Carboniferous sediments, have eroded away to form valleys while the basalt filled valleys have resisted erosion to form the ridge tops. Evidence supporting this model includes the occurrence of a Cenozoic gravel deposit underlying basalt on the Williams Range (Bruce 1995). Such deposits (or deep leads) are indicative of buried valleys. Similar Cenozoic gravel deposits now situated on ridges have been reported by Mason (1982) from beneath the basalt from the western part of the lava field. Erosion has dramatically altered the Barrington landscape since the Paleocene/Eocene basaltic Proc. Linn. Soc. N.S.W., 132, 2011 M.C. BRUCE eruptions in the southern part of the lava field. This has led to some impressive rock formations such as columnar basalt to be found on and around places like Mount Allyn and Mount Lumeah. The plateau and all its distinctive natural features is a post basalt erosional surface (Galloway 1967) and possibly represents remnants of scarp retreat of the Great Dividing Range (Ollier 1982). Rubies, sapphires and zircons of gem quality and rare secondary minerals are associated with the lava field (Sutherland and Graham 2003). CONCLUSIONS Geological mapping in the southern Barrington Tops lava field has identified 33 basaltic flows correlated from several localities over 630m of topographic relief. This matches the number of flows recorded by Mason (1982) albeit over only 430m of relief, in the Prospero Trigonometric station sequence in the north-western lava field. In the southern region, flows 10-20m thick are separated by either an agglomerate or sub-horizontal palaeosols. . Petrography and whole-rock geochemistry reveal basanites dominate the lava sequence interspersed with alkali basalts, pyroxene-phyric basalts (ankaramites) and tholeiites. Alkali gabbros form intrusions near the top of the sequence and represent conduits for surface lavas. Major and trace elements of the alkaline magmas along with fractionation/ accumulation controlled modelling are consistent with the magmas being co-genetic and chiefly controlled by a low pressure olivine+ plagioclase mineral assemblage. Clinopyroxene did fractionate without greatly affecting magma compositions. Incompatible trace elements are consistent with low degree melting of an enriched mantle source (OIJB-type), although there is evidence of entrained amphibole-enriched sub-continental lithospheric mantle. A third, more depleted mantle component, may have contributed to the tholeiites. These conclusions are consistent with alkali basalts analysed from the northern part of the lava field (O’Reilly and Zhang 1995; Sutherland and Fanning 2001) where basalt generation was linked to thermal anomalies in the mantle causing low degree asthenospheric and lithospheric melting. These melts rose up relatively quickly where they largely pooled in high crustal level magma chambers, fractionated olivine + clinopyroxene and accumulated plagioclase before venting. The presence of peridotite xenoliths (Powell and O’Reilly 2007) suggests that at least some of the magma ascended directly from the upper mantle, sampling lower crustal material (gabbroic enclave) enroute to the surface. Proc. Linn. Soc. N.S.W., 132, 2011 ACKNOWLEDGEMENTS Research for this project was undertaken in 1995 as partial requirements of a Bachelor of Science (Hons) Degree with the then School of Earth Sciences at Macquarie University, Sydney. Ian Percival and Trevor Green are thanked as principal supervisors whose guidance and discussions proved invaluable. Norm Pearson and Carol Lawson provided assistance with geochemical analysis. Jim Starling (NSW National Parks and Wildlife) and Mike Prima (NSW State Forests) gave permission to visit and sample localities at Barrington. I am grateful for reviews by Lin Sutherland and Suzanne O’Reilly that improved the final version of the manuscript. Published with the permission of the Acting Director, Geological Survey of New South Wales. REFERENCES Benson, W.N. (1912). Preliminary note on the nepheline- bearing rocks of the Liverpool and Mount Royal Ranges. Journal and Proceedings of the Royal Society of New South Wales 45, 176-186. Binns, R.A., Duggan, M.B. and Wilkinson, J.F.G. (1970). High pressure megacrysts in alkaline lavas from northeastern New South Wales. American Journal of Science 69, 132-168. Bruce, M.C. (1995). The petrology, geochemistry and origin of the Tertiary basalt from the southern Barrington Tops lava field, New South Wales. BSc (Hons) thesis, Macquarie University, Sydney. Crawford, A.J., Stevens, B.P.J. and Fanning, M. (1997). Geochemistry and tectonic setting of some Neoproterozoic and Early Cambrian volcanics in western New South Wales. Australian Journal of Earth Science 44, 831-852. Eggins, S. (1984). Geology and geochemistry of the Barrington Tops Batholith. BSc (Hons) thesis, University of New South Wales, Sydney. Galloway, R.W. (1967). Pre-basalt, sub-basalt and post- basalt surfaces of the Hunter Valley, New South Wales. In ‘Landform studies from Australia and New Guinea’ (Eds J.N. Jennings and J.A. Mabutt) pp 293-313. (Australian National University Press, Canberra). Green, T.H. (1992). Petrology and geochemistry of basaltic rocks from the Balleny Isles, Antarctica. Australian Journal of Earth Sciences 39, 603-617. Hoernle, K., White, J.D.L., van der Bogaard, P., Hauff, F., Coombs, D.S., Werner, R., Timm, C., Garbe- Schonberg, D., Reay, A. and Cooper, A.F. (2006). Cenozoic intraplate volcanism on New Zealand: Upwelling induced by lithospheric removal. Earth and Planetary Science Letters 248, 350-367. Le Bas, M.J., Le Maitre, R.W., Streckeisen, A. and Zanettin, B. (1986). A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27, 745-750. 67 SOUTHERN BARRINGTON TOPS LAVA FIELD Mason, D.R. (1982). Stratigraphy of western parts of the Barrington Tops Tertiary volcanic field. In “New England Geology: Voisey Symposium 1982’ (Eds P.G. Flood and B. Runnegar) pp 133-139. (University of New England, Armidale, Australia). Mason, D.R. (1985). Polybaric crystallisation of clinopyroxene in ankaramites of the Barrington Tops Tertiary volcanic field, New South Wales. Geological Society of Australia, New South Wales Division Publication 1, 87-105. Mason, D.R. and Kavalieris, I. (1984). A preliminary note on the Barrington Tops Granodiorite, New South Wales, Australia. American Mineralogist 71, 1314- 1321. Middlemost, E.A.K. (1985). ‘Magmas and magmatic rocks’. (Longman, London). Norrish, K. and Hutton, J.T. (1969). An accurate X- ray spectrographic method for the analysis of a wide range of geological samples. Geochimica Cosmochimica Acta 33, 431-453. Norrish, K. and Chappell, B.W. (1977). X-ray fluorescence spectrometry. In “Physical methods in determinative mineralogy’ (Ed J. Zussman) pp 201-272. (Academic Press, London) Ollier, C.D. (1982). The Great Escarpment of eastern Australia: tectonic and geomorphic significance. Journal of the Geological Society of Australia 29, 36-43. O’Reilly, S.Y. and Zhang, M. (1995). Geochemical characteristics of lava-field basalts from eastern Australia and inferred sources: connections with the subcontinental lithospheric mantle? Contributions to Mineralogy and Petrology 121, 148-170. Pain, C.F. (1983). Geomorphology of the Barrington Tops area, New South Wales. Journal of the Geological Society of Australia 30, 187-194. Pearce, T.H. (1968). A contribution to the theory of variation diagrams. Contributions to Mineralogy and Petrology 19, 142-157. Powell, W. and O’Reilly, S. (2007). Metasomatism and sulphide mobility in lithospheric mantle beneath eastern Australia: Implications for mantle Re-Os chronology. Lithos 94, 132-147. Roberts, D.L., Sutherland F.L., Hollis J.D., Kennewell P. and Graham, I.T. (2004). Gemstone characteristics North East Barrington Plateau, NSW. Journal and Proceedings of the Royal Society of New South Wales 137, 99-122. Rollinson, H. (1993). “Using geochemical data: evaluation, presentation, interpretation’. (Longman Group Limited, UK). Russell, J.K. and Nicholls, J. (1988). Analysis of petrological hypotheses with Pearce element ratios. Contributions to Mineralogy and Petrology 99, 25- 35. Sun, S.S. and McDonough, W.F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle compositions and processes. In ‘Magmatism in ocean basins’ (Eds A.D. Saunders and M.J. Norry) pp 313-345. (Geological Society of 68 London Special Publication 42). Sutherland, F.L. and Fanning, C.M. (2001). Gem-bearing basaltic volcanism, Barrington, New South Wales: Cenozoic evolution, based on basalt K-Ar ages and zircon fission track and U-Pb isotope dating. Australian Journal of Earth Sciences 48, 22\-237. Sutherland, L. and Graham, I. (2003). ‘Geology of Barrington Tops Plateau: Its rocks, minerals and gemstones, New South Wales, Australia’. (The Australian Museum Society, Sydney). Wellman, P. (1989). Upper mantle, crust and geophysical volcanology of eastern Australia. In “Intraplate Volcanism in eastern Australia and New Zealand’ (Ed R.W. Johnson) pp 29-38. (Cambridge University Press, Cambridge). Wellman, P., McElhinney, M.W. and McDougall, I. (1969). On the polar wander path for Australia during the Cainozoic. Geophysical Journal of the Royal Astronomical Society 18, 371-395. Proc. Linn. Soc. N.S.W., 132, 2011 Table A1 Major and trace element XRF analyses, southern Barrington lava field. M.C. BRUCE Sample MUS5361 MU55365 MUS5366 MU55369 MU55372 MU55378 MU55381 MUS5382 MUS5383 Barrington Tops 1:25k GR.492538 GR.500539 GR.553660 GR.502533 GR.553522 GR.528540 GR.520553 GR.524550 GR.553518 9133-1-N Pyroxene- Rock type Alkali basalt phyricbasalt Basanite Alkaligabbro Basanite Basanite Alkalibasalt Alkalibasalt Basanite wt% Sid2 47.44 45.25 43.92 48.9 44.16 43.63 45.02 44.95 44.15 TiO2 1.97 2.55 2.31 2.3 2.53 2.44 2.36 2.46 2.35 Al203 14.49 16.01 14.27 16.38 14.54 13.25 14.82 14.87 14.7 Fe203 1.93 1.85 1.91 1.85 1.91 1.96 1.92 1.98 1.88 FeO 9.63 9.28 9.55 9.23 9.55 9.8 9.64 9.9 9.42 MnO 0.17 0.18 0.19 0.16 0.19 0.19 0.18 0.18 0.2 MgO 10.44 7.31 41.14 6.28 10.95 13.63 10.59 10.49 10.86 CaO 9.8 11.73 11.47 9.67 11.5 10.39 10.29 10.06 10.94 Na20 2.72 3.32 3.38 3.28 2.98 2.97 3.05 3 2.6 K20 1.11 1.77 0.68 1.24 0.85 0.64 1.42 1.48 1.34 P205 0.39 0.96 0.99 0.5 1.07 0.73 0.82 0.82 0.85 Total 100.09 100.21 99.81 99.79 100.23 99.53 100.11 100.19 99.29 ppm Ba 409 712 567 381 580 427 417 430 681 Rb 20 33 10 17 22 16 20 22 31 Sr 634 1017 1450 666 1146 917 889 989 41141 Y 20 27 24 25 23 22 24 25 23 Zr 122 226 194 158 246 188 231 233 177 Nb 46 115 79 53 103 85 79 81 79 Th 1 7 4 1 6 2 5 4 5 Pb 5 3 4 5 3 6 6 4 5 Ga 19 23 19 24 20 7 20 19 20 Zn 90 76 80 84 80 84 78 81 58 Cu 43 62 54 43 51 39 61 62 47 Ni 171 82 181 52 177 320 204 208 178 Vv 228 240 245 219 243 248 245 233 238 Cr 351 112 325 84 337 327 372 326 303 Sample MU55385 MU55386 MU55393 MUS55397 MU55401 MU55409 MU55411 MU55418 MU55424 Barrington Tops 1:25k GR.563513 GR.555492 GR.565516 GR.555490 GR.519466 GR.518458 GR.518444 GR.560480 GR.526441 9133-1-N Rock type Basanite Alkali basalt Basanite Trachybasalt Basanite Basanite Basanite Tholeiite Alkali basalt wt% Si02 44.85 45.01 42.87 46.1 43.33 43.01 43.59 49.15 45.17 TiO2 2.28 2.19 2.15 2.65 2.25 2.45 2.38 1.49 1.61 Al203 14.41 14.8 13.73 15.97 14.1 13.87 13.35 16.42 13.49 Fe203 1.93 1.99 1.82 1.9 1.84 1.81 1.82 2.09 1.92 FeO 9.67 9.97 9.1 9.49 9.15 9.06 9.1 10.44 9.56 MnO 0.17 0.18 0.19 0.18 0.18 0.19 0.18 0.16 0.2 MgO 41.23 11.31 11.64 7.99 11.28 12.12 12.32 6.33 13.22 CaO 10.58 10.53 11.56 9.19 11.1 12.19 11.96 10.03 10.24 Na20 2,59 2.5 3.67 3.32 3.31 2.78 2.72 2.77 2.69 K20 0.81 0.95 0.96 1.78 0.93 0.76 0.81 0.52 0.75 P205 0.66 0.6 1.23 0.88 1.25 1.22 1.2 0.32 0.75 Total 99.18 100.03 98.92 99.45 98.72 99.46 99.43 99.74 99.6 ppm Ba 442 355 771 606 687 663 680 146 517 Rb 9 11 12 22 15 13 16 6 12 Sr 843 768 1322 1328 1330 1292 1413 412 776 Y 23 20 24 27 23 25 24 22 23 Zr 165 146 219 324 251 242 202 80 125 Nb 62 55 109 121 100 123 98 19 51 Th 3 2 10 3 6 4 4 0 5 Pb 4 5 6 2 3 5 5 3 4 Ga 18 19 17 19 15 15 16 21 14 Zn 75 80 77 83 74 73 73 80 78 Cu 64 67 45 33 43 53 42 71 65 Ni 224 213 252 107 206 198 239 208 388 Vv 212 222 225 195 205 256 245 192 210 Cr 312 318 372 115 361 367 366 294 542 Proc. Linn. Soc. N.S.W., 132, 2011 69 nT i gee ee 0 ie — — 1 PO A SS aT eT ; —— Pi oes wie. headers pnw Euty a ee oF et rT ee Pie SLM. eh as ae PUD him Pi i el! Oe ere) pe =e a - me ee ee er 22 ——— re Til 4 pe nn ovens — qin “as avauvd Ce ee Veen teed Sa a x apn tai ——————— eae thang Sm 9 a) ya as ania e a 4 ie ie og = ‘ Pi» oad rez: Fl a: ibhe we ee Cort, me a Mh Lee BOT ist e. am z > Pia Taye P : im 4 e ia ihe Be * g ‘ m; am yee i a 5 41 al Pt ie at as ‘i od a ie oll Pit i , oe ¢*9 ee ' @ - * a eee * oF ae 2 oe i, i ear i ee 8 — er | “os a as i Oe eS e Pine eG Fa aot Ao Shee ze. 4 a wert . ee . - . - rte wees — ome — if jena oie ae pea Mi, Rita Neca TUR: Cael ‘mer Easeenae 7 argrnenin a een my Ua Dg ae ge ; ral — - me : ‘Ae Salt Me vein ste, agri? mid wi: nee ore armen ceo | eRe — a ~yrnee . _ Dieeehe . ; _ — I ol “rte eames ee fnaenoeit onl ctaied pee tise et Kania ti C0 ee

. Proc. Linn. Soc. N.S.W., 132, 2011 S. MEAKIN Behr H.J., Behr K and Watkins J.J. (2000). Cretaceous microbes — producer of black opal at Lightning Ridge, NSW, Australia. Jn: Skilbeck C.G. and Hubble T.C.T. (editors). Understanding Planet Earth: Searching for a Sustainable Future. Abstracts of the 15" Australian Geological Convention, University of Technology, Sydney. Geological Society of Australia Abstracts 59: 28. Brammall, J. and Smith, E.T. (2007). Historic sites of the Three Mile opal field, Lightning Ridge. Unpublished report to NSW Department of Primary Industries. Lightning Ridge Opal & Fossil Centre Inc. Brook, B., Gillespie, R. and Martin, P. (2006). Megafauna mix-up — fresh evidence indicates that humans were responsible for the extinction of Australia’s megafauna. Australasian Science 27(5):35-37. Burger, D. (1980). Palynology of the Lower Cretaceous Surat Basin. Bureau of Mineral Resources, Geology and Geophysics Bulletin 189, 1-106. Burton, G.R. (2010). Angledool 1:250 000 Geological Sheet SH/55—7, 2nd edition. Provisional map. Geological Survey of New South Wales, Maitland, NSW. Burton, G.R. (in press). Angledool 1:250 000 Geological Sheet SH/55—7, 2nd edition. Explanatory Notes. Geological Survey of New South Wales, Maitland, NSW. Byres, J.G. (1977). Notes on the Rolling Downs Group in the Milparinka, White Cliffs and Angledool 1:250 000 sheet areas. Geological Survey of New South Wales, Report GS1977/005. Carpenter, R.J., Goodwin M.P., Hill, R.S. and Kanold, K. (in litt.). Silcrete plant fossils from Lightning Ridge, New South Wales: new evidence for climate change and monsoon elements in the Australian Cenozoic. Darragh, P J, Gaskin, A J & Sanders, J V 1976, ‘Opals’, Scientific American vol. 234(4), pp. 84-95. Dettman, M.E., Molnar, R.E., Douglas, J.G., Burger, D., Fielding, C., Clifford, H.T., Francis, J., Jell, P., Wade, M., Rich, P.V., Pledge, N., Kemp, A. and Rozefelds, A. (1992). Australian Cretaceous terrestrial faunas and floras: biostratigraphic and biogeographic implications. Cretaceous Research 13, 207-262. Environment Protection and Heritage Council (2009). 19" Meeting of the EPHC, Communiqué. 6p. http:/Avww.productstewardship.asn.au/documents/ EPHCCommunique05112009-1.pdf. Viewed 22/02/2011. Field, J. and Dodson, J.R. (1999). Late Pleistocene megafauna and archaeology from Cuddie Springs, southeastern Australia. Proceedings of the Prehistoric Society 65, 275-301. Field, J., Wroe, S. and Fullagar, R. (2006). BLITZKREIG: Fact and fiction at Cuddie Springs. Australasian Science 27(6): 28. Flannery, T. F., Archer, M., Rich, T. H. and Jones, R. (1995). A new family of monotremes from the Cretaceous of Australia. Nature 377: 418-420. Hamulton-Bruce, R.J. and Kear, P. (2010). A possible succineid land snail from the Lower Cretaceous non- Proc. Linn. Soc. N.S.W., 132, 2011 marine deposits of the Griman Creek Formation at Lightning Ridge, New South Wales. Alcheringa 34, 325-331. Hamilton-Bruce, R. J., Smith, B. J. and Gowlett-Holmes, K. L. (2002). Descriptions of a new genus and two new species of viviparid snails (Mollusca: Gastropoda: Viviparidae) from the Early Cretaceous (middle-late Albian) Griman Creek Formation of Lightning Ridge, northern New South Wales. Records of the South Australian Museum 35(2): 193-203. Hocknull, S.A., (2000). Mesozoic freshwater and estuarine bivalves from Australia. Memoirs of the Queensland Museum 45, 405-426. Idriess, Ion L. (1944). Lightning Ridge. Angus & Robertson Classics, Australia. Kear, B. P. (2006). A new fossil unionoid bivalve from the lower Cretaceous non-marine deposits of Lightning Ridge, eastern Australia. Abstracts. Riversleigh 2006 Symposium. University of NSW, Sydney. Kemp, A. and Molnar, R. E. (1981). Neoceratodus forsteri from the lower Cretaceous of New South Wales. Journal of Palaeontology. 55: 211-217. Martin H.A. (1980). Stratigraphic palynology from shallow bores in the Namoi River and Gwydir River Valleys, north central New South Wales. Journal and Proceedings, Royal Society of New South Wales, 113, 81-87. Martin H.A. (1981). Stratigraphic palynology of the Castlereagh River Valley, New South Wales. Journal and Proceedings, Royal Society of New South Wales, 114, 77-84. McLoughlin, S. and Kear, B.P. (2010). The Australasian Cretaceous scene. A/cheringa 34, 197-203. Molnar, R.E. (1980). Procoelous crocodile from the Lower Cretaceous of Lightning Ridge, NSW. Memoirs of the Queensland Museum 20: 65-75. Molnar, R. E. (1999). Avian tibiotarsi from the early Cretaceous of Lightning Ridge, New South Wales. In: Proc. 2" Gondwanan Dinosaur Symposium (eds.) Y. Tomida, T. H. Rich and P. Vickers-Rich. National Science Museum Monographs 15, 197-209, Japan. Molnar, R.E. (2010). Taphonomic observations on eastern Australian Cretaceous sauropods. Alcheringa 34, 421-429. Molnar, R. E. and Galton, P. M., (1986). Hypsilophodontid dinosaurs from Lightning Ridge, New South Wales, Australia. Geobios 19: 231-239 Molnar, R. E. and Willis, P. M. (2001). New crocodilian material from the Early Cretaceous Griman Creek Formation, at Lightning Ridge, New South Wales. In: Crocodilian Biology and Evolution (ed.) C. Grigg- Gordon. Surrey-Beatty and Sons London. Morgan, R. (1984). Palynology. In: Contributions to the Geology of the Great Australian Basin in New South Wales. (eds.) J.M. Hawke and J. N. Cramsie. Geological Survey of NSW Bulletin 31, NSW Department of Mineral Resources. Musser, A. M. (2005). Investigations into the evolution of Australian mammals with a focus on Monotremata. Unpubl. PhD thesis. Univ. of NSW, Sydney 291 pp. V2 LIGHTNING RIDGE GEODIVERSITY Narran Lakes Ecosystem Project undated. Fact sheet 3: The First Settlers. Murray—Darling Basin Commission, Monash University, Griffith University, water CRC, University of Canberra . Viewed 15 February 2011 at http://www.canberra.edu.au/centres/ narran/docs/resources/factsheets/NFS_3.pdf>. NSW Department of Mineral Resources (2000). Mining at Lightning Ridge. Minfact 95. Pecover, S R (1996). A new genetic model for the origin of opal in the Great Australian Basin. In: Mesozoic Geology of the Eastern Australia Plate Conference, Geological Society of Australia Inc, Extended Abstracts No. 43, 450-454. Pecover, S R (1999). A new syntectonic model of origin to explain the formation of opal veins (‘seams’ & ‘nobbies’), breccia pipes (‘blows’) and faults (‘slides’) at Lightning Ridge. First National Opal Mining Symposium, Extended Abstracts. 30-31 March, Lightning Ridge. Predavec, M., Claridge J., Blackwell J. and France, L. (2004). Opal mining within the Narran—Warrambool Reserve, Lightning Ridge. Review of Environmental Factors. Report prepared for the NSW Department of Mineral Resources. Parson Brinckerhoff Australia Pty Ltd. Rey, P.F., Glass-van der Beek, I. and Davies, P.J. (2003). Some remarks about the formation seam opal. Third National Opal Mining Symposium 2003, Extended Abstracts. 22-25 September, Quilpie. Rich, T.H. and Vickers-Rich, P. (1994). Neoceratopsians and Ornithomimosaurs: dinosaurs of Gondwana origin? National Geographic Research and Exploration 10(1):129-131. Scheibnerova, V. (1974). Cretaceous foraminifera of the Great Australian Basin. Geological Survey of NSW Memoir Palaeontology 17. NSW Department of Mineral Resources. Scheibnerova, V. (1984). Micropalaeontology. In: Contributions to the Geology of the Great Australian Basin in NSW (eds.) J. M. Hawke and J. N. Cramsie. Geological Survey of NSW Bulletin 31, NSW Department of Mineral Resources. Smith, E.T. (2007). White dirt, tailing heaps and lively ground — Early Cretaceous fossil sites of Lightning Ridge. Unpublished report to NSW Department of Primary Industries. Lightning Ridge Opal & Fossil Centre Inc. Smith, E.T. (2009). Terrestrial and freshwater turtles of Early Cretaceous Australia. Unpublished PhD thesis, University of NSW, Sydney, 390pp. Smith, E. T. (2010). Early Cretaceous chelids from Lightning Ridge, New South Wales. Alcheringa 34, 375-384. Smith, E. and Smith, R. (1999). Black opal fossils of Lightning Ridge — treasures from the Rainbow Billabong. Kangaroo Press. 112 pp. Taylor, G. (1976). The Barwon River, New South Wales — a study of basin fill by a low gradient stream in a semi-arid climate. PhD thesis, Australian National University, Canberra (unpublished). 80 Taylor, G. (1978). A brief Cainozoic history of the Upper Darling Basin, Royal Society of Victoria, Proceedings 90(1), pp 53-59. Tourism Research Australia (2008). Tourism profiles for Local Government Areas in regional Australia. Walgett Shire. Tourism Australia brochure. Viewed 2 December 2010 . Turner, S. (2006a). Forging a Geopark Network in the Australasia—Pacific region. In: 2°’ UNESCO International Geoparks 2006 Conference, Belfast. Abstracts. Geological Survey of Ireland, p. 106. Turner, S. (2006b). Promoting UNESCO Global Geoparks for sustainable development in the Australian—Pacific region. Alcheringa Special Issue 1, pp. 351-365. Watkins, J. J. (1985). Future prospects for opal mining in the Lightning Ridge region, Geological Survey of New South Wales, Report GS1985/119. Watkins, J.J and Meakin, N.S. (1996). Nyngan and Walgett 1:250 000 geological sheets SH/55-15 & SH/55-11: Explanatory Notes. Geological Survey of New South Wales, Sydney. White, M.E. (1986). The Greening of Gondwana. Reed Books, Sydney. Proc. Linn. Soc. N.S.W., 132, 2011 S. MEAKIN APPENDIX. List of Cretaceous fossil fauna of the Griman Creek Formation, Lightning Ridge, from Smith (2007, 2009, 2010), with contributions from other sources (p.c. = pers. comm.). Taxon Source Chlorophyceae Charophyta indet. taxon Dr Adriana Garcia p.c; Henk Godthelp p.c. Foraminifera Hyperammina sp Scheibnerova 1984 Ramulina tetrahedralis Ludbrook 1966 Scheibnerova 1984 Radiolaria ? radiolarian Scheibnerova 1984 Polychaeta indet. taxon Mollusca Pelecypoda Alaythyria jagueti Newton 1915 Megalovirgus wintonensis Hocknull 1997 Hyridella macmichaeli Hocknull 1997 Hyridella (Protohyridella) goondiwindiensis Hocknull 1997 Palaeohyridella godthelpi Hocknull 2000 Coocrania hamiltonbrucei Kear 2006 large ?hyriid sphaeriid ‘tellen’ or nut shell corbiculid — river pea shell strongly ridged, subcircular unioid clam with spines, narrow rippled margins Gastropoda Albianopalin benkeari Hamilton-Bruce et al. 2002 Albianopalin lizsmithae Hamilton-Bruce et al. 2002 Notopala sp. Hamilton-Bruce et al. 2002 Melanoides godthelpi Hamilton-Bruce et al. 2004 Fretacaeles gautae Hamilton-Bruce and Kear 2006 Suratia marilynae Hamilton-Bruce and Kear 2010 Crustacea Decapoda freshwater crayfish — indet. taxon Anura frog — indet. taxon Dr Mike Tyler and Henk Godthelp p.c. Pisces Chondrichthyes small shark cf /surus or Cretolamna small shark cf /surus or \ Cretolamna Actinopterygia indet. taxa x 4 Dr Sue Turner pers. comm. — Teleostei aspidorhynchid cf Richmondichthys sweeti Etheridge and Smith Woodward 1891 Proc. Linn. Soc. N.S.W., 132, 2011 81 Ichthyosauria Sauropterygia Testudines - turtles Crocodilia Pterosauria Dinosauria Mammalia 82 LIGHTNING RIDGE GEODIVERSITY Dipnoi — lungfish Pliosauria Plesiosauria Chelidae Testudines indet. Ornithopoda Sauropodo- morpha Theropoda Aves Synapsida Monotremata Taxon Source freshwater eel — indet. taxon Dr Peter Forey and Dr Tom Rich p.c.. Ceratodus wollastoni Chapman 1914 Kemp and Molnar 1981 Ceratodus diutinus Kemp 1993 Kemp 1993 Neoceratodus forsteri Kreftt 1870 Kemp and Molnar 1981 ichthyosaur - indet. taxon Dr Benjamin Kear p.c. ?Leptocleidid pliosaur Dr Benjamin Kear p.c. elamosaurid plesiosaur - indet. taxon Dr Benjamin Kear p.c. indeterminate chelid pleurodires x 2 taxa Smith 2010 meiolaniid-like taxon 1 - ‘Spook’s Turtle’ Smith 2009 meiolaniid-like taxon 2 - ‘Sunflash Turtle’ Smith 2009 Crocodylus selaslophensis Etheridge 1917 crocodile — ziphodont Molnar and Willis 2001 crocodile — conical tooth form Molnar and Willis 2001 pterosaur - indet. taxon Henk Godthelp p.c. stegosaurid Dr Benjamin Kear p.c. Muttaburrasaurus sp Molnar 1991, 1996 Fulgurotherium australe von Huene 1932 (Molnar and Galton 1986) Atlascopcosaurus loadsi Rich and Rich 1989 Leallynasaurus sp? Rich and Rich 1989 very large hypsilophodontid very large hypsilophodontid indeterminate sauropods x 2 - ‘spoon tooth’ form, ‘sharp tooth’ form small ?prosauropod Rapator ornitholestoides von Huene 1932 ? alvarezsaurid or ceratosaurid - very large form dromaeosaurid cf. Velociraptor Henk Godthelp p.c. ornithomimosaurid Henk Godthelp p.c. ? spinosaurid Dr Benjamin Kear p.c. unidentified ornithoracines - two taxa Molnar 1999 unidentified ?synapsid Clemens et al. 2003 Steropodontidae - Steropodon galmani Archer et al. 1985 Kollikodontidae - Kollikodon ritchiei Flannery et al. 1995 2Ornithorhynchidae - up to 3 unidentified Smith 2009 taxa Proc. Linn. Soc. N.S.W., 132, 2011 Wee Jasper—Lake Burrinjuck Fossil Fish Sites: Scientific Background to National Heritage Nomination GAVIN C. YOUNG Research School of Earth Sciences, The Australian National University, Canberra ACT 0200, Australia (Gavin. Young@anu.edu.au) Young, G.C. (2011). Wee Jasper—Lake Burrinjuck fossil fish sites: scientific background to National Heritage Nomination. Proceedings of the Linnean Society of New South Wales 132, 83-107. The ~5 km thick Burrinjuck Devonian sedimentary sequence records environmental change from a volcanic terrain with deep lake deposits (oldest), through a tropical reef marine ecosystem, to river and lake deposits (youngest). Numerous fossil horizons document evolutionary change through the final stage of terrestrialization of the earth’s biota. Exceptional exposures of Devonian tropical reefs in the Wee Jasper valley, with limestones washed completely clean by the waters of Lake Burrinjuck, have produced the world’s oldest known coral reef fish assemblage. Including associated invertebrates, the faunal list stands at some 266 fossil genera. Burrinjuck produced five key fossil fish specimens used in the 1940s in London to develop the acetic acid technique for extracting bone from calcareous rock (now standard in laboratories throughout the world). Recognizing the uniquely preserved early vertebrate braincase structures, the British Museum (Natural History) mounted two collecting expeditions to Burrinjuck (1955, 1963), when some 560 specimens were removed to London. Repatriation of type specimens is a future issue. The largest collection of Burrinjuck early vertebrate braincase material is housed at the Australian National University in Canberra; at least 70 fossil fish species represents biodiversity unequalled at any other Devonian fossil fish locality. Fossil site protection for the Burrinjuck area was the basis for a recent nomination for National Heritage listing. Long- term protection of natural history collections in the National Capital as part of Australia’s scientific heritage is a related issue of concern. Manuscript received 30 November 2010; accepted for publication 16 March 2011. KEYWORDS: Burrinjuck, Cavan, coral reefs, Devonian fishes, vertebrate braincase, Wee Jasper. INTRODUCTION The Devonian Period (~418-360 million years ago), known as the ‘Age of Fishes’, was the geological period when the early jawed vertebrates underwent their first great evolutionary radiation. This included not only abundant and diverse fishes in all habitable aquatic environments, but also an evolutionary expansion of our ancestors (the first four-legged land animals) into an entirely new terrestrial environment, made habitable by the rise of land plants including the first forests during the Devonian Period. One of the NSW geological heritage sites (Taemas-Cavan) described by Percival (1985, pp. 30-33) represents the Devonian Period, and occupies the southeastern arm of Lake Burrinjuck, about 50 km NW of the National Capital in southeastern Australia (Fig. 1). In that area, Early Devonian marine limestones display spectacular folding, and richly fossiliferous shell beds such as are exposed at the ‘Shearsby’s Wallpaper’ protected site. Taemas-Cavan is the easternmost of two main areas of outcrop of Devonian limestones around Burrinjuck Dam (Fig. 1B). The western outcrop, surrounding the village of Wee Jasper in the valley of the Goodradigbee River, is separated by the Narrangullen anticline, where underlying older units (Mountain Creek Volcanics; Kirawin Formation; Sugarloaf Creek Formation) are exposed. Research on the geology and especially the vertebrate palaeontology of the Burrinjuck area has given it national and international significance. In March, 2010 the author lodged a nomination for part of the western outcrop of the Burrinjuck limestones to be included on the National Heritage List (areas labelled (1), Fig. 1C). Significant cave and karst structures documented by A. Spate form part of that nomination but are not dealt with further here (for details contact A. Spate, Optimum Karst Management). In this paper I summarise the geological and palaeontological significance of the Devonian sequence, with special BURRINJUCK AREA FOSSIL FISH SITES Lake Burrinjuck e Cave | ang if QWade island Burrinjuck °."..\\ an Garevah FM a [, *_] Hatchery Creek Gp er Banks SER LOWER DEVONIAN 7 Roads ; oT ral Murrumbidgee Group ; (G9 Black Range Volcanic SILURIAN [=] Undifferentiated Picarbera Cc Taemas Bridge Cave Island Cavan Burrinjuck Dam \ | Cooradigbee fe Carey's Cave 1 i i sal Seen ey Wee Jasper Figure 1. A. General localities for the two Australian limestone reef fos- sil fish assemblages from the Devonian Period (~418-360 million years ago): Gogo, WA (GG) and Burrinjuck, NSW (BJ). B. Geological map of the Burrinjuck area showing the two outcrops of Murrumbidgee Group limestones: Good Hope — Taemas - Cavan in the east, valley of the Goodradigbee River in the west. C. Lake Burrinjuck (shaded) on the Murrumbidgee River, with location of the area nominated for heritage listing in March 2010 (labelled (1), indicated by solid out- line). Proposed future Stage II nomination (labelled (I1)) indicated by dashed outline, including the NSW geological heritage area of Percival (1985, map 9) indicated by cross-hatching. Other localities mentioned in the text also labelled (‘Cave Island = pre-dam ‘Cave Flat’). 84 Proc. reference to the remarkable fossil vertebrate remains that have been extracted from the Burrinjuck limestones. A brief summary of recent results of ongoing research is presented. HISTORY OF SCIENTIFIC INVESTIGATION The limestones of the Goodradigbee and Murrumbidgee valleys were doubtless well known to the indigenous population for many thousands of years, because of their interesting rock formations and caves. The limestone outcrops were first noted by Europeans as early as 1824 (by Hume and Hovell), and in 1836 fossil corals were collected from the area by the explorer Thomas Mitchell (Mitchell 1838). From 1848, the ‘Father of Australian geology’ Rev. W.B. Clarke made many collections in the area (Clarke 1860, 1878), which he sent overseas for expert determination (de Koninck 1877) to confirm the Devonian age for the limestones. Bennett (1860, p. 158) mentioned visits to the “‘Gudarigby Caverns’, apparently the limestone caves at Cave Flat near the junction of the Murrumbidgee and Goodradigbee rivers. Etheridge (1889) reported on a visit to these caves, where a spectacular Thylacinus skull was collected, and displayed for many years in the fossil gallery of the Australian Museum, Sydney (see Fig. 4A). Etheridge recognized the exceptional fossil content of the limestones, which he described in the following terms (1889, p. 36): ‘The Murrumbidgee limestone is ... crammed with fossils, especially corals. As a display of these beautiful organisms in natural section I have never seen its equal. Large faces of limestone ... may be seen, with the weathered Linn. Soc. N.S.W., 132, 2011 G.C. YOUNG corals ... standing out in relief and in section also. Many of these masses of coral, particularly those of Stromatopora and Favosites, are as much as 4 feet in diameter.’ A few years later Etheridge also visited the ‘caves at Goodravale, Goodradigbee River’ (now Carey’s Cave, Wee Jasper valley, included in the heritage nominated area; see Fig. 1C), where cave deposits produced jaw remains of the marsupial lion Thylacoleo (Etheridge 1892). Etheridge (1906) then reported the first discovery of a Devonian lungfish from Burrinjuck, at that time the oldest known representative of the Dipnoi (see below). Harper (1909) conducted a geological mapping survey preliminary to the proposal to dam the Murrumbidgee River in the Burrinjuck area. After construction of Burrinjuck Dam (1912-17) the area became more widely visited, and was aregular destination for geology student excursions from the University of Sydney. The subject matter of Professor T.W. Edgeworth David’s first Australian publication (1882) was the geology and palaeontology of this area, and Dr Ida Browne did detailed stratigraphy - and produced the first geological maps (Browne 1954, 1959). Her 1959 paper, including her widely used geological map of the Taemas area, was part of a publication to mark the centenary of the birth of Professor Edgeworth David. In recent decades Burrinjuck has been a focus for geology student excursions from many universities, and especially the ANU in Canberra, because of proximity and research interest. With more frequent droughts in recent years the rock exposures and fossil sites along the shores and on the bed of Lake Burrinjuck are now often accessible for extended periods. BURRINJUCK DEVONIAN SEQUENCE The Devonian Period lasted for some 60 million years. However, the exceptionally thick Burrinjuck Devonian sequence, comprising some 5 km of sedimentary strata overlying an equivalent or greater thickness of volcanics (Fig. 2), was mainly deposited during the early part of the Devonian Period (see below). A general observation is a strong cyclicity evident on a larger scale through some 3 km of Stratigraphic thickness comprising the uppermost Hatchery Creek Group, and the Murrumbidgee Group limestones. In the limestones this is manifested as more recessive units comprising shale/limestone interbeds alternating with more massive limestones as constituent members of the Taemas Limestone (Browne 1959, Young 1969, Pedder et al. 1970). Proc. Linn. Soc. N.S.W., 132, 2011 In the Hatchery Creek Group fining-upward cycles occur throughout the succession, but with generally finer and less thick cycles higher in the sequence (Hunt and Young 2010). Google images to the east of Wee Jasper suggest a downward continuation of this cyclicity into the underlying Sugarloaf Creek Formation. The phenomenon could relate to orbital forcing causing regularity in climatic fluctuations (Hunt and Young, submitted). Elsewhere (e.g. Middle Devonian of Scotland, Late Devonian Munster Basin of southwest Ireland) smaller scale (36, 55 m) and larger scale (130 m) sedimentary cycles have been attributed to 100 Ka and 412 Ka Milankovitch Cycles respectively, with somewhat lesser thicknesses (8, 40 m sedimentary cycles) attributed to 21 Ka and 100 Ka Milankovitch Cycles in the largely lacustrine Orcadian Basin (Kelly 1992, Kelly and Sadler 1995, Marshall et al. 2007). This aspect of the Burrinjuck Devonian sequence has not been researched in any detail. The last comprehensive accounts of the Devonian stratigraphy were by Owen and Wyborn (1979) for the Brindabella 1:100 000 geological map, and by Cramsie et al. (1978) for the northern part of the outcrop on the Yass 1:100 000 geological map. The following stratigraphic summary (oldest to youngest) relies heavily on Owen and Wyborn’s (1979) explanatory notes (microfiche portion, now converted to pdf). Mountain Creek Volcanics The name was first published by Joplin et al. (1953). Some authors grouped this unit with the overlying Kirawin and Sugarloaf Creek Formations as the ‘Black Range Group’, but Owen and Wyborn (1979) considered these three units too dissimilar to be grouped together. The upper part of the Mountain Creek Volcanics in the Cavan area comprises rhyolites and tuffs deposited in a terrestrial environment. Estimated total thickness farther south is 5000-8000 m (Owen and Wyborn 1979, p. M190). The Mountain Creek Volcanics are considered to be entirely Devonian (probably mostly Lochkovian) in age, on the assumption that rhyolites at Mount Bowning on the Yass 1:100 000 sheet are equivalent (Link 1970). These rhyolites overlie lowermost Devonian strata containing the early Lochkovian conodont Jcriodus woschmidti (Link and Druce 1972; see Pogson 2009 for updated comment on the conodonts). Kirawin Formation This black shale/mudstone deposit forms a poorly exposed outcrop 0.5 to 4 km wide, in an arcuate 35 km beltacross the Narrangullen anticline. The outcrop thins 85 BURRINJUCK AREA FOSSIL FISH SITES zs 22/9 s A Oo s5|/a~=- a &| Ww ‘ ws <= | ao ee | —=-+---—/9. 221E Si\